Homework answers / question archive / You will select a research topic, generate testable hypotheses, review relevant literature, describe participant selection procedures, identify data collection methods, describe the proposed quantitative research design, address potential ethical problems, and describe limitations of your research proposal

You will select a research topic, generate testable hypotheses, review relevant literature, describe participant selection procedures, identify data collection methods, describe the proposed quantitative research design, address potential ethical problems, and describe limitations of your research proposal

Health Science

Share With

You will select a research topic, generate testable hypotheses, review relevant literature, describe participant selection procedures, identify data collection methods, describe the proposed quantitative research design, address potential ethical problems, and describe limitations of your research proposal.

This assignment is the fourth step toward completing your final research proposal in Week 6. In a two to three page paper, in addition to the Title Page and References Page, answer the following. Please use the Kallet (2004) and Krosnick (1999) articles for further guidance on how to prepare the methods section and how to construct a survey.

Data Collection Measure: Choose one of the following to measure your data:

1. Design a survey (10-15 items): This should be ready for administration, so include clear instructions about how respondents should take, complete, and return the survey. Pay careful attention to issues of design: question ordering, question phrasing, closed- or open-ended questions, overall convenience and attractiveness of the questionnaire, etc.
2. Select a systematic observation technique: Describe the setting for gathering your data, write instructions for observations to be made, explain how you will code and/or analyze these data to measure your variables, etc.

Be sure to include an APA formatted title page and references page.

COVID-19, Small Business Owners, and Racial Inequality Robert Fairlie Robert Fairlie is a professor of economics at the University of California, Santa Cruz and an NBER research associate affiliated with the Economics of Education and Productivity, Innovation, and Entrepreneurship Programs. He is a regular participant in the Entrepreneurship Working Group and Economics of Education meetings, and plans on participating in the new Race and Stratification in the Economy Working Group. Fairlie’s research interests include entrepreneurship, education, racial inequality, information technology, labor economics, and immigration. Recent research projects explore questions around causes and consequences of racial inequality, barriers to business creation and growth, whether technology helps students, constraints in higher education, whether there have been disproportionate impacts of COVID-19 by race and gender, and water conservation policy. Fairlie received his PhD and MA from Northwestern University and BA with honors from Stanford University. He has held visiting positions at Stanford, Yale University, UC Berkeley, and Australian National University. He has received funding for his research from numerous government agencies and foundations and has testified to the US Senate, US House of Representatives, the Department of Treasury, and the California State Assembly regarding the findings of his research, and received a joint resolution of appreciation from the California legislature. He is regularly contacted by major news media to comment on economic, small business, inequality and policy issues. 12 NBER Reporter • No. 4, December 2020 The widespread closing of businesses in the United States and around the world due to the coronavirus has been unprecedented. Stores, factories, and many other businesses have closed as a result of policy mandates, downward demand shifts, health concerns, or other factors. Although many have reopened since social distancing restrictions were relaxed, the revenues lost from the closures, the limited scale of current reopenings, and the potential for further closures in the future may lead to a wave of permanent small business closures with disproportionate impacts by race, gender, and nativity. In several recent papers, I examine the impacts of COVID19 on small business owners, using timely microdata from the Current Population Survey (CPS) and administrative data from the Small Business Administration. These new papers build on my longstanding research agenda on entrepreneurship, racial inequality, and small business policy. This summary reviews selected papers from both recent and earlier work. Early Stages of the Pandemic On March 19, 2020, the state of California imposed shelter-in-place restrictions, with New York State following the next day. By early April, most states had imposed social distancing restrictions that closed “nonessential” businesses and added to consumer health concerns in the emerging pandemic. Using CPS microdata, I examine how COVID-19 impacted small business owners in mid-April 2020, the first month to capture these changes.1 Figure 1 shows that the number of working business owners plummeted from 15 million in February 2020 to 11.7 million in April 2020, the largest drop ever; the entire Great Recession only resulted in a drop of 5 percent. Even incorporated business owners, who tend to be more stable and growth-oriented than Number of Active Business Owners in the US, 2005–2020 Millions 18 16 14 12 10 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Source: Fairlie R W. NBER Working Paper 27309, and published as "The impact of COVID-19 on small business owners: The first three months after social-distancing restrictions", Journal of Economics & Management Strategy, 29(4), 2020, pp 727–40. Figure 1 unincorporated ownthe population. In conNumber of Active Business Owners by Ethnicity before and during COVID-19 ers, experienced a drop trast, EIDL loans and in work activity of 20 advances, in both numMillions percent from February ber and amounts, were 2.0 to April 2020. provided positively to Losses for busiminority communities. 1.5 nesses owned by 02/2020 04/2020 05/2020 06/2020 Ties to Broader, women, racial minorities, and immiLong-Term Racial 1.0 grants were espeInequality cially severe [Figure In earlier re­search, 2]. African Americans 0.5 I explore the link experienced the largbetween racial inequalest losses: a 41 perity in business outcent drop in the num0.0 comes and broader ber of active business African-American Latinx Asian racial inequality. owners. Latinx busiResearch on earnness owners also expeSource: Fairlie R W. NBER Working Paper 27309, and published as "The impact of COVID-19 on small business owners: The first three months after social-distancing restrictions", Journal of Economics & Management Strategy, 29(4), 2020, pp 727–40. ings inequality almost rienced major losses: exclusively focuses on 32 percent. Immigrant Figure 2 the wage and salary secbusiness owners suftor and ignores the other fered a 36 percent drop, major way to make a living — owning a and female business owners 25 percent. Policy Response to COVID-19 business. Ten percent of the workforce, or Concentrations of female, Black, Latinx, Given the severity of the pandemic, 12 million people, own a business rather and Asian businesses in industries hit hard by the pandemic, such as personal the federal government provided more than holding a wage or salaried job. These services, partly explain why the losses were financial assistance to small businesses owners hold a disproportionate amount of higher for these groups than the national than ever previously seen. The largest pro- total wealth and create jobs for others. Racial disparities in business formaaverage. Extending the analysis into the grams providing funds to small businesses second and third months following wide- were the $660 billion Paycheck Protection tion raise concerns about lost economic spread shelter-in-place restrictions — May Program (PPP) and the $220 billion efficiency. If minority entrepreneurs face and June 2020 — business owner activity Economic Injury Disaster Loan (EIDL) liquidity constraints, discrimination, or partially rebounded, but the dispropor- program. One of the goals stated in the other barriers to creating new businesses tionate impacts from COVID-19 by gen- Coronavirus Aid, Relief, and Economic or expanding current ones, there will be der, race, and immigrant status lingered. Security (CARES) Act, which included efficiency losses in the economy. Barriers African Americans continued to experi- the PPP and EIDL programs, was to pri- to entry and expansion are potentially ence the largest losses, with 26 percent of oritize assistance to underserved markets costly to productivity and local job creation, especially as minorities represent a formerly active business owners still not and disadvantaged business owners. But did the PPP and EIDL programs growing share of the population. reactivated in May and 19 percent not In a series of papers, I use various active in June. Job losses were also higher get disbursed to minority communities? Frank Fossen and I explore this question datasets to study the causes of racial and for minority workers.2 Overall, these early estimates of the using administrative data on the universe ethnic disparities in business ownership, impact of COVID-19 on small busi- of PPP loans, EIDL loans, and EIDL formation, and outcomes, focusing on the nesses indicate that losses were spread advances.4 We generally find a slightly constraints that limit productivity and across demographic groups and types of positive relationship between PPP loan cause inefficiencies in the economy. Work business — no group was immune — but receipt per business and the minority with Alicia Robb draws on confidensome groups were hit harder than others. share of the population. There is some evi- tial, restricted-access, business-level data Although there is no way to determine dence that the first round of funds was dis- from the US Census Bureau to explore at present whether these business clo- proportionately disbursed to nonminor- why Asian American-owned firms persures will be permanent, each additional ity communities and that the second was form well in comparison to White-owned month of inactivity has an impact on the disproportionately disbursed to minor- businesses, while Black-owned firms typirevenues, profits, and employees of these ity communities. Focusing on PPP loan cally do not.5 We find differential access businesses, and on their likelihood of amounts per employee, we find a negative to financial capital to be the largest facrelationship with the minority share of tor. Family business experience also plays ever reopening.3 NBER Reporter • No. 4, December 2020 13 a role in explaining differences in out- els of education and wealth explain the ership and find evidence of crowd-out comes. In more recent work, I exam- entire gap between Mexican immigrants between immigrant and native owners.10 ine potential barriers created by human and non-Latinx Whites in business forcapital, wealth, demographic, geographic, mation rates; together with language abil- Small Business Policies and industry constraints for each group ity, these factors explain nearly the entire Governments and donors spend bilusing CPS and American Community gap in business income. Legal status repSurvey data.6 I find that low levels of resents an additional barrier for Mexican lions of dollars subsidizing entrepreneurwealth contribute to lower rates of Black immigrants. ship training and development programs and Latinx business ownership, and Using census microdata from the around the world. Arguments for subthat high levels of wealth increase Asian United States, Canada, and the United sidizing training are manifold, and span business ownership rates. Low levels of Kingdom, Harry Krashinsky, Julie theories of allocative and/or redistribueducation contribute to lower business Zissimopoulos and I provide the first tive frictions in credit, labor, insurance, income for Blacks and Latinx, and high comparative examination of the educa- and human capital markets. Dean Karlan, levels of education increase Asian busi- tion levels, business ownership, and busi- Jonathan Zinman and I explore the effecness income. The Black, tiveness of entrepreLatinx, and Asian popuneurship training Average Annual Sales Per Small Business, by Ethnicity, 2012 lations are all relatively programs by working young compared to the with US Department $600,000 White population; this of Labor data from also contributes to lower the largest random $546,176 $500,000 business ownership rates experiment ever conin these groups. ducted evaluating $400,000 Using confidential entrepreneurship $364,717 and restricted-access training.11 After con$300,000 trolling for selection panel data from the into training, we find Kauffman Firm Survey, $200,000 that entrepreneurship along with matched training has a sizable administrative data on $100,000 $143,271 short-term impact credit scores, Robb, $58,119 on increasing busiDavid Robinson and $0 ness ownership and I explore disparities White African-American Latinx Asian reducing unemployin capital use between Source: Fairlie R W. NBER Working Paper 27309, and published as "The impact of COVID-19 on small business owners: The first ment, but no effect Black- and Whitethree months after social-distancing restrictions", Journal of Economics & Management Strategy, 29(4), 2020, pp 727–40. on business ownowned startups.7 We find that Black-owned ership or any busiFigure 3 startups start smaller ness outcome such as and stay smaller over the first eight years ness performance of Asian immigrants.9 sales, exit rates, profits, or employment in of their existence. Black startups face We find that business ownership rates of the medium and long term. more difficulty in raising external capi- Asian immigrants in the United States and Policymakers have sought to improve tal, especially external debt. We find that Canada are similar to the national aver- success among minority business owndisparities in creditworthiness constrain ages, and in the UK they are substantially ers. In the United States, for example, Black entrepreneurs; perceptions of treat- higher than the national average and the although they are sometimes controverment by banks also hold them back. Black highest among the three countries. Asian sial, a variety of federal, state, and local entrepreneurs apply for loans less often immigrants even from the same source government programs offer contractthan White entrepreneurs largely because country are generally much more edu- ing goals, price discounts, and loans to they expect to be denied credit, even cated in the United States than in Canada businesses owned by minorities, women, when they have a good credit history and or the United Kingdom. Although there and other groups that are historically in settings where strong local banks favor are many institutional, structural, and his- underrepresented among business owntorical differences between the countries ers. Aaron Chatterji, Kenneth Chay and new business development. Christopher Woodruff and I study that might be responsible, one possibility I examine the effectiveness of affirmawhy Mexican-American entrepreneur- is that the higher returns to education in tive action contracting programs for busiship is low in the United States even the United States result in a more selective nesses owned by African Americans by though self-employment rates are very immigrant pool. Bruce Meyer and I study using the staggered introduction of these high in Mexico.8 We find that low lev- how groups interact in business own- contracting programs across cities in the 14 NBER Reporter • No. 4, December 2020 1980s.12 Black business ownership rates increased significantly after program initiation. On average, the Black-White gap fell 3 percentage points. Black gains were concentrated in industries heavily affected by contracting programs, and they mostly benefited those who were better educated. 1 “The Impact of COVID-19 on Small Business Owners: Evidence of Early-Stage Losses from the April 2020 Current Population Survey,” Fairlie R. NBER Working Paper 27309, June 2020. Published as “The Impact of COVID-19 on Small Business Owners: Evidence from the First Three Months after Social-Distancing Restrictions,” Journal of Economics & Management Strategy 29(4), Winter 2020, pp. 727–740. Return to Text 2 “The Impacts of COVID-19 on Minority Unemployment: First Evidence from April 2020 CPS Microdata,” Fairlie R, Couch K, Xu H. NBER Working Paper 27246, May 2020. Published as “Early Evidence of the Impacts of COVID-19 on Minority Unemployment,” Journal of Public Economics 192, December 2020. Return to Text 3 “The Impact of COVID-19 on Small Business Owners: Evidence of Early-Stage Losses from the April 2020 Current Population Survey,” Fairlie R. NBER Working Paper 27309, June 2020. The findings from this paper were cited in deliberations of the US Senate Committee on Small Business and Entrepreneurship and used to justify policy responses providing assistance to minority and small businesses during the pandemic. “Cardin, Cantwell, Schumer, Booker, Cortez Masto & Harris Introduce Legislation to Invest in Minority-Owned Businesses,” press release, office of Cardin B. “Governor Newsom Signs Bills to Support Small Businesses Grappling with Impact of COVID-19 Pandemic, Bolster Economic Recovery,” press release, State of California, Office of the Governor. Return to Text 4 “Did the $660 Billion Paycheck Protection Program and $220 Billion Economic Injury Disaster Loan Program Get Disbursed to Minority Communities in the Early Stages of COVID-19?” Fairlie R, Fossen F. NBER Working Paper 28321. Return to Text 5 Race and Entrepreneurial Success: Black-, Asian-, and White-Owned Businesses in the United States, Fairlie R, Robb A. Cambridge: MIT Press, 2008. “Why Are Black-Owned Businesses Less Successful than White-Owned Businesses? The Role of Families, Inheritances, and Business Human Capital,” Fairlie R, Robb A. Journal of Labor Economics 25(2), April 2007, pp. 289–323. Return to Text 6 “Racial Inequality in Business Ownership and Income,” Fairlie R. Oxford Review of Economic Policy 34(4), Winter 2018, pp. 597–614. Return to Text 7 “Black and White: Access to Capital among Minority-Owned Startups,” Fairlie R, Robb A, Robinson D. NBER Working Paper 28154, November 2020. Return to Text 8 “Mexican Entrepreneurship: A Comparison of Self-Employment in Mexico and the United States,” Fairlie R, Woodruff C. NBER Working Paper 11527, August 2005, and in Mexican Immigration to the United States, Borjas G, editor. Chicago: University of Chicago Press, 2007. Return to Text 9 “The International Asian Business Success Story? A Comparison of Chinese, Indian and Other Asian Businesses in the United States, Canada and United Kingdom,” Fairlie R, Zissimopoulos J, Krashinsky H. In International Differences in Entrepreneurship, Lerner J, Schoar A, editors, pp. 179–208. Chicago: University of Chicago Press, 2010. Return to Text 10 “The Effect of Immigration on Native Self-Employment,” Fairlie R, Meyer B. NBER Working Paper 7561, February 2000, and Journal of Labor Economics 21(3), July 2003, pp. 619–650. Return to Text 11 “Behind the GATE Experiment: Evidence on Effects of and Rationales for Subsidized Entrepreneurship Training,” Fairlie R, Karlan D, Zinman J. NBER Working Paper 17804, August 2012, and American Economic Journal: Economic Policy 7(2), May 2015, pp. 125–161. Return to Text 12 “The Impact of City Contracting Set-Asides on Black Self-Employment and Employment,” Chatterji A, Chay K, Fairlie R. NBER Working Paper 18884, March 2013, and Journal of Labor Economics 32(3), July 2014, pp. 507–561. Return to Text NBER Reporter • No. 4, December 2020 15 Copyright of NBER Reporter is the property of National Bureau of Economic Research, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. Innovation Strategies of Mature Resilient Businesses during the Covid-19 crisis Adrien Lecossier Arts et Métiers ParisTech, LAMPA 2, Bd du Ronceray, 49000 Angers, France adrien.lecossier@ensam.eu Marc Pallot Arts et Métiers ParisTech, LAMPA 2, Bd du Ronceray, 49000 Angers, France marc.pallot@ensam.eu Abstract—The current sanitary crisis all over the world is unprecedented and affects all areas of activities. While companies have been investing in innovation strategies for years, crisis periods are very often conducive to downscaling R&D investment and realizing incremental innovations. However, the COVID-19 crisis caused the emergence of new multidimensional stakes summarize in the “Monde d’Après” movement. This exploratory study proposes to analyze the innovation strategy of mature companies applied during the COVID-19 sanitary crisis in order to understand if they apply specific strategies according to the "unprecedented" aspect of this crisis. The results based on the analysis of collected lessons learned show that most of companies mainly focus on incremental innovation strategies during this crisis. necessary business activities than emergent companies and start-ups [4]. Keywords—Innovation crisis strategies, Mature companies, Established companies, Covid-19, Innovation strategies. At the societal level, this unprecedented crisis has brought about a will to question the current functioning of society with citizen consultations on the "Monde d’Après" (Next World) [7], [8]. For example, in France, the citizen consultation carried out from April 10 to May 25, 2020 on the Make.org platform made possible to collect 20,000 ideas submitted by 165,000 participants and 1.7 million votes [9]. The question asked in the face of the crisis caused by the coronavirus epidemic was: "How do we all invent the next world?". 14 main categories of ideas were defined by compiling ideas which received at least 80% of the votes "agree" and at least 37% of the mentions "favorite" or "realistic". As an example, the first six categories are: (1) promoting local consumption and proximity circuits; (2) move towards an agricultural alternative; (3) limit the production of waste, in particular plastic packaging, and encourage recycling; (4) Relocate certain strategic economic sectors in France and in Europe; (5) Rethinking education favouriting humans and the environment; (6) Put the environment and the social at the heart of public policies and taxation [9]. I. INTRODUCTION The sanitary crisis being experienced today is one of the most serious that the world has ever known. Billions of people have been confined to their homes for about 2 months, with outside excursions authorized only for basic and vital activities. Those mobility restrictions prevent people to go to work and consume normally. Consequently, it has a direct impact on the economy as it downgrades at the same time both the supply and the demand. From an unprecedented sanitary crisis is born an economic crisis which also promises to be unprecedented [1]. According to a very recent study, industrial production has decreased by about 13.5%, seasonally adjusted retail sales were down 21%, car sales fell 92%, and restaurant sales dropped by almost 95% in China, where COVID-19 first struck [2]. Therefore, to face this sanitary crisis, many companies were forced to reduce or even stop their business, either under regulatory and sanitary constraints or because of a lack of employees at work and orders/demand. This particular restricted production capacity of businesses has led, for example, to a massive use of teleworking (part-time working). In France, from mid April 2020, more than 821K companies have already put in place governmental partial unemployment measures that represented over 2 million workers impacted [3]. At the corporate level, such a crisis requires adoption of a crisis management approach. Crisis management is the set of activities dedicated to perform: (a) prevention; (b) preparation; (c) response; (d) recovery [5]. Often, adopting crisis management means focusing on protecting the present while the future receives less attention. Indeed, most of companies implement plant savings strategies reducing thus their workforce, CAPEX and investments in R&D and innovation [6]. People around the world are currently thinking about the “Monde d’Après” which could influence the way the resilient and mature companies are dealing with the crisis. This challenging societal and economic situation forces companies to seek new types of crisis strategies beyond reductions of workforce, CAPEX, and investments in R&D and innovation. In front of all crises, businesses are not equal. It is known that established corporations and mature companies are more inclined to protect employment and the continuation of © IEEE 2020. This article is free to access and download, along with rights for full text and data mining, re-use and analysis 2020 IEEE International Conference on Engineering, Technology and Innovation (ICE/ITMC) II. CHALLENGES CREATED BY THE COVID-19 CRISIS A. Prior research on challenges facing mature companies A mature industrial company seeks to maintain a deterministic and profitable model that it has built over time, called dominant design [10]. For this, it essentially creates incremental innovations that reduce costs of existing products and/or substantially improves their performance in order to sustain their market share [11], [12]. In contrast, and in order to survive on long terms, an industrial company must create radical innovations [13], [14]. The creation of radical innovation is touchy to manage for many practitioners [15]. According to Rajapathirana and Hui’s study, the innovation capability is the most required component for developing effective radical innovation outcomes within a firm. It requires understanding and identifying the future customer needs, expectations and potential of adoption [16]. Such an innovation capability requires developing an innovation culture to enable creation of new ideas and transform them into successful radical innovations [17]. To acquire this capability, more and more companies have been actively, and sometimes for years, facing the challenge to structure their upstream innovation process, with specific processes, methodologies, tools and practices [18]. B. New challenges caused by the Covid-19 As mentioned in the introduction, a crisis such as Covid-19 makes the activities addressing the future business receive far less attention, especially radical innovation activities [6]. Indeed, as observed, the largest applied strategy for a company in time of crisis, relies on creating incremental innovations; it consists in reducing the costs of existing products and / or substantially improving their performance in order to sustain their market share and growth margin [11], [12]. However, despite the Covid-19 crisis, mature businesses are still faced with the need to renew themselves. In some cases, this need may be amplified or even accelerated depending on the crisis impact on the company's economic activities. Therefore, a mature company should define a specific innovation crisis strategy to ensure that the available resources are well serving its resilience and renewal capability. III. PRIOR RESEARCH ON SPECIFIC INNOVATION STRATEGIES In order to help businesses design an innovation approach adapted to their specific context, a multidimensional vision of the innovation typologies was proposed [19]. This tool offers the possibility for a company to position and define innovation strategies according to a market and technological context it is facing. The tool is structured using innovation intensities and typologies presented in the next sections. A. Innovation intensities 1) Radical innovation: A radical innovation, also called disruptive innovation, is often used to describe the creation of a new product causing disruption in the market. It fits well with Schumpeter's "creative destruction" for whom radical innovation is the destruction of the existing systems by novelty [2]. The work of Garcia and Calantone supports this definition in more detail and characterizes radical innovation as an innovation that causes market and technology discontinuities at macro and micro levels [15]. More generally, a radical innovation allows improving the performance of the dominant design. However, this increase must be 5 to 10 times greater than the existing system, or disruptive to the dominant design, to be considered radical [20], [21]. 2) Incremental innovation: An incremental innovation is closer to optimization, improvement, and minor add on. B. Innovation typologies 1) Product innovation: A product innovation is a new or improved good or service that differs significantly from the firm’s previous goods or services, and has been introduced on the market [22]. When radical, it modifies the dominant design which causes the appearance of a discontinuity at the level of the company and of the whole market. ? IPT#1: Goods include tangible objects and some knowledge-capturing products over which ownership rights can be established and whose ownership can be transferred through market transactions [23]. ? IPT#2: Services are intangible activities that are produced and consumed simultaneously and that change the conditions (e.g. physical, psychological, etc.) of users. The engagement of users through their time, availability, attention, transmission of information, or effort is often a necessary condition that leads to the co-production of services by users and the firm. The attributes or experience of a service can therefore depend on the input of users. Services can also include some knowledge- capturing products [23]. Whether relating to goods or services, product innovations are based on a combination of modular, architectural changes and often process innovations [19]. 2) Process innovation: A business process innovation is a new or improved business process for one or more business functions that differs significantly from the firm’s previous business processes and has been brought into use in the firm [22]. The Oslo manual lists six functional categories of business process innovations: ? IP#1: Production of goods or services innovations including all the innovations that transform inputs into goods or services, including engineering innovations and related technical testing, analysis and certification activities to support production. ? IP#2: Distribution and logistics innovations including transportation and service delivery innovations; warehousing innovations; order processing innovations. ? IP#3: Marketing and sales including marketing methods innovations (product promotion and placement, packaging of products, direct marketing, exhibitions and fairs, market research and other activities to develop new markets); pricing strategies and methods innovations; sales and after-sales innovations (help desks other customer support and customer relationship activities). ? IP#4: Information and communication systems innovations including hardware and software 2020 IEEE International Conference on Engineering, Technology and Innovation (ICE/ITMC) innovations; data processing and database innovations; maintenance and repair innovations; web-hosting and other computer-related information innovations. ? IP#5: Administration and management innovations including strategic and general business management innovations (cross-functional decision-making, organizing work responsibilities); corporate governance innovations (legal, planning and public relations); accounting, bookkeeping, auditing, payments and other financial or insurance innovations; human resources management innovations (training and education, staff recruitment, workplace organization, provision of temporary personnel, payroll management, health and medical support); procurement innovation ; managing external relationships innovations. ? IP#6: Product and business process development innovations including all innovations concerning the way to scope, identify, develop, or adapt products or a firm's business processes. What sets business process innovations apart from an incremental product innovations is that their scope is limited to procedure, process or process type results without touching a user outside the organization [22]. the number of the two types of innovation slows down and tends to stabilize [22]. C. Dominant design To establish links between the typologies of product and process innovation, the work of Abernathy and Utterback is of interest, which shows that product and process innovations follow one another in a temporal sequence of three phases on the scale of the industry [24] (Fig 1). ? If so, do mature companies really favor the realization of incremental innovation? Brockhoff, Hauschildt and Chakrabarti established links between the dominant design and innovation intensities [27]. Their works shows that the first two phases are periods of radical change where product and process innovations are introduced into an emerging environment, while the specific phase is a period of incremental changes within a more harmonious whole. IV. MAIN ISSUES During a crisis, a mature company must work to: (1) increase its resilience; and (2) develop its capacity to create radical innovations. Consequently, the company efforts can include product as well as process radical and incremental innovations. It more globally means that, in order to answer these two main stakes, a mature company must realize all types of innovation, a very difficult and unrealistic task in crisis-time. This therefore raises questions about several points: ? Is there a crisis innovation strategy specific to mature companies? ? Are mature companies implementing innovation strategies that remain sufficiently exploratory to continue to find opportunities for renewal? These are the questions dealt with in the next section. Number of production innovation 2 Number of business process innovation 1 1 2 3 3 t Fig. 1. Evolution of the product and business process innovation t Utterback explains that the first phase is characterized by the fact that the number of product innovations is greater than the number of process innovations [23]. He points out that the proliferation of product innovations eventually ends with the emergence of a dominant model commonly called dominant design. It is during this phase that the whole innovation cycle starts, an idea for a product is first designed to realize a product innovation, and leading to the need to create business process innovations [25], [26]. The second phase, called the transition phase, is distinguished by the decrease in the number of product innovations and by the growth in the number of process innovations [21]. It is at the start of this phase that the emergence of a dominant design arrives to reduce the variety of product innovation, and that development efforts will gradually focus on producing the dominant design more efficiently. During the third phase, which the authors call "specific phase", V. RESEARCH APPROACH A. Objectives and research questions The main objective of this research is to understand if a particular crisis-driven innovation strategy, based on a combination of different innovation typologies, does really exist. An additional goal is to evaluate if the application of this strategy can support a company to become more resilient and renew itself during the crisis-time. The main research question to verify the objectives is: ? What types of innovation strategy have companies preferred to apply during a crisis and why? An answer to the research question was pursued by focusing on the analysis of ideas that have emerged during the crisis in an industrial group composed of 28 daughter companies that did not apply a specific crisis innovation strategy. This analysis allows to precisely identify what innovation strategy is naturally emerging during a crisis when no dedicated strategy is predefined. B. Methodology Collected lessons learned after two months of crisis were analyzed. The lessons learned were shared among different department managers and directors of 28 Business Units of an 2020 IEEE International Conference on Engineering, Technology and Innovation (ICE/ITMC) international industrial group. The collection of lessons learned was achieved in using an ideation platform. Contributors had to synthesize their lessons learned through few sentences, the use of a picture, or keywords. Contributors also had to assign one or more structural improvement items that correspond, for instance, to the different innovation typologies of their contributions. Finally, the used approach for the study can be considered as a deductive and descriptive research approach [28]. C. Data collection and sampling Respondents (n=102) shared in total 124 lessons learned using the idea-sharing platform. Each of them included: (1) Qualitative comments for describing the lessons learned (title, description, keywords), (2) Qualitative categories for defining their types through the use of multiple choice. Among the 124shared lessons learned, 113 were selected for the analysis, because some of them were only considered as encouragement messages. D. Analysis method The innovation typologies grid enabled classification of the 113 selected lessons learned. For each lessons learned, a reading of the title and keywords was done before the reading of the detailed description, the latter containing, on average, 867 characters. VI. ANALYSIS AND FINDING In order to better qualify the innovation strategies that companies apply during the crisis, a classification of the lessons learned that are considered as innovations or ideas of innovation, shared by the 113 contributors of the 28 business units, is presented. To evaluate how the crisis efforts are exactly distributed, the shared lessons learned were assigned to the one or more innovation typologies. Among the 113 lessons learned analyzed, the main efforts are concentrated on the realization of incremental innovations (81%). This result is detailed using a distribution analysis for which more than one innovation typology can be attributed to one lesson learned. As shown in Table I: (a) approximately 78% of these incremental innovations refer to ideas of administrative, management and human resources innovation; (b) 21% to ideas of innovations which relate to information and communication systems; (c) 10% to developments in product and business process; (d) 7% to developments in marketing, sales or pricing methods; (e) and less than 3% concern distribution innovations. Only 4% of the shared incremental innovations refer to service innovations, belonging to the category of product innovations. TABLE I. DISTRIBUTION OF THE SHARED LESSONS LEARNED USING PRODUCT AND PROCESS INNOVATIONS CATEGORISATION Total (n) IPT #1 IPT #2 IP# 1 IP# 2 IP# 3 IP# 4 IP# 5 IP# 6 5 11 3 3 12 20 76 20 Rad. (n) Inc. (n) Total (%) Rad. (%) Inc. (%) Distri. Rad. (%) Distri. Inc. (%) 5 7 3 0 6 1 4 11 0 4 0 3 6 19 72 9 4 10 3 3 11 18 67 18 4 6 3 0 5 1 4 10 0 4 0 3 5 17 64 8 24 33 14 0 29 5 19 52 0 4 0 3 7 21 78 10 Not indicated in the Table I, about 19% of the 113 learned lessons analyzed are concentrated on the realization of radical innovations. Table I, indicates 52% of these radical innovations refer to radical adaptation of products or business processes, 29% to developments in marketing, sales or pricing methods, 19% refer to ideas of administrative, management and human resources innovation, 14% to developments in process supporting the creation of radical product innovations, and 5% concern information and communication systems. Curiously, no radical learned lessons concerned a distribution innovation idea. In contrast with the incremental lessons learned, 33% of the radical ones concern service innovation and 24% goods innovation. More globally, 57% of radical lessons learned belong to the product innovation category while only 4% belong to the incremental category. Overall, and beyond the degrees of innovation, most (67%) shared lessons learned concern ideas of administrative, management and human resources innovations VII. DISCUSSION The results of the study show that the innovation strategy applied by companies is mainly a process innovation strategy (89%). This behaviour can be explained by the fact that the Covid-19 crisis required a review of the organizational operations of companies. The innovations made the companies test new ways of working. Therefore, the information and communication systems innovations, representing 20% of the lessons learned, are higher than others because they mainly represent the IT innovations supporting the organisational ones (e.g. home office, teleconference). However, beyond organizational and information systems innovations, two other types of process innovations remain above product innovations. This study’s results confirm the conclusion of other studies, that process innovations are preferred by mature companies during crisis because they lower production cost [6]. It also confirms that process innovations better fit in a “bank-oriented” financial system [29]. Hence, it reflects how mature companies are resilient in the way they serve shareholders interest. In fact, mature companies are 2020 IEEE International Conference on Engineering, Technology and Innovation (ICE/ITMC) keener on investing in this type of activity with respect to a riskier product innovation, especially during an economic crisis where uncertainty is much higher. [10] [11] VIII. CONCLUSION AND FUTURE WORK After analysing 113 lessons learned from 28 companies in a group, this study shows that the most applied innovation strategy is that of incremental process innovation. It therefore appears that these incremental changes initiated within organizations during the crisis may ultimately bring about new ways of working. The crisis did not appear to be ‘The opportunity’ for these companies to initiate new disruptive projects based on eventual new business models challenging the dominant design. However, for double-checking whether industrial companies consider the Covid-19 crisis as an opportunity to reduce their costs or renew their business, it could be interesting to carry out a survey on a larger scale within several sectors. In conclusion, this study demonstrates that the “Monde d’Après” has not yet appeared within the mature companies. As with other crises prior, the Covid-19 crisis was an opportunity to optimise the process and reduce costs. However, perhaps if new citizen initiatives are supported by governmental power, mature companies could create and apply new disruptive strategy in order to serve the new “Monde d’Après” paradigm. This last point opens new research perspectives particularly focused on the way researchers could help practitioners to accomplish this change. REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] N. Fernandes, “Economic effects of coronavirus outbreak ( COVID19 ) on the world economy Nuno Fernandes Full Professor of Finance IESE Business School Spain,” SSRN Electron. Journal, ISSN 15565068, Elsevier BV, pp. 0–29, 2020. M. McKee and D. Stuckler, “If the world fails to protect the economy, COVID-19 will damage health not just now but also in the future,” Nat. Med., 2020. A. Tonnelier and R. Besse Desmoulières, “Chômage partiel la facture du coronavirus encore largement sous estimée,” Le Monde, 2020. [Online]. Available: https://www.lemonde.fr/politique/article/2020/04/22/chomagepartiel-la-facture-du-coronavirus-encore-largement-sousestimee_6037456_823448.html. G. S. Walsh and J. A. Cunningham, “Business failure and entrepreneurship: Emergence, evolution and future research,” Foundations and Trends in Entrepreneurship. 2016. N. Altay and W. G. Green, “OR/MS research in disaster operations management,” Eur. J. Oper. Res., vol. 175, no. 1, pp. 475–493, 2005. C. Flammer and I. Ioannou, “To Save or To Invest? Strategic Management During the Financial Crisis Caroline,” SSRN Electron. J., pp. 1–41, 2018. L. Monde, “Le monde d’après : « Il est temps de ne pas reprendre comme avant »,” 2020. [Online]. Available: https://www.lemonde.fr/idees/article/2020/05/07/le-monde-d-apresil-est-temps-de-ne-pas-reprendre-commeavant_6038917_3232.html. [Accessed: 07-May-2020]. Politizr, “Le Monde d’Après...Vous !,” 2020. [Online]. Available: https://www.politizr.com/groupes/le-monde-d-apres-vous/votremonde-d-apres. [Accessed: 05-May-2020]. Make.org, “Crise Covid-19 : Comment inventer tous ensemble le monde d’après ?,” 2020. [Online]. Available: [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] https://about.make.org/le-monde-dapres. [Accessed: 25-Apr-2020]. J. Utterback, “Mastering the Dynamics of Innovation: how companies can seize opportunities in the face of technological change,” Harvard Bus. Sch. Press, 1994. M. E.Porter, “Competitive Advantage - Creating and Sustaining Superior Performance.” p. 580, 1985. A. Rolstadås, B. Henriksen, and D. O’Sullivan, “The Innovation Process,” in Manufacturing Outsourcing: A Knowledge Perspective, London: Springer London, 2012, pp. 129–144. C. Christensen, “The innovator’s dilemma,” Harvard Bus. Sch. Press., 1997. Gaia Rubera and A. H. Kirca, “Firm Innovativeness and Its Performance Outcomes: A Meta-Analytic Review and Theoretical Integration,” J. Mark., vol. 76, no. 3, pp. 130–147, 2012. A. Lecossier, M. Pallot, P. Crubleau, and S. Richir, “Towards radical innovations in a mature company: An empirical study on the UX-FFE model,” Artif. Intell. Eng. Des. Anal. Manuf. AIEDAM, vol. 33, no. 2, pp. 172–187, 2019. R. P. J. Rajapathirana and Y. Hui, “Relationship between innovation capability, innovation type, and firm performance,” J. Innov. Knowl., pp. 1–16, 2017. S. F. Slater, J. J. Mohr, and S. Sengupta, “Radical Product Innovation Capability: Literature Review, Synthesis, and Illustrative Research Propositions,” J. Prod. Innov. Manag., vol. 31, no. 3, pp. 552–566, 2014. J. Nicholas, “An Investigation into the Practices and Underlying Factors during the Fuzzy Front End of Radical Innovation,” no. April, 2014. A. Lecossier, P. Crubleau, F. Goux-Baudiment, and S. Richir, “Une vision multidimensionnelle des typologies d’innovation pour identifier et concevoir une démarche d’innovation,” in CONFERE 16, 2016. G. C. O’Connor and M. Rice, “Opportunity Recognition and Breakthrough Innovation in Large Established Firms,” Calif. Manage. Rev., vol. 43, no. 2, pp. 95–116, 2001. R. Garcia and R. Calantone, “A critical look at technological innovation typology and innovativeness: a literature review,” J. Prod. Innov. Manag., vol. 19, pp. 110–132, 2002. OECD/Eurostat, Oslo Manual 2018: Guidelines for Collecting, Reporting and Using Data on Innovation, 4th Edition, The Measurement of Scientific, Technological and Innovation Activities. 2018. System of National Accounts 2008. 2009. W. Abernathy and J. Utterback, “Patterns of industrial innovation,” Technol. Rev., no. June-July, pp. 40–47, 1978. R. G. Cooper, Winning at new products : accelerating the process from idea to launch, Cambridge : Third Edition Basic Books, 2001., Third edit., vol. 33, no. 3. Cambridge, 2001. Q. Zhang and W. J. Doll, “The fuzzy front end and success of new product development: a causal model,” Eur. J. Innov. Manag., vol. 4, no. 2, pp. 95–112, 2001. J. H. and A. K. C. K. Brockhoff, The dynamics of Innovation, Springer. Berlin: Springer Berlin Heidelberg, 1999. H. Dumez, Méthodologie de la recherche qualitative, Vuibert. Paris, 2015. E. Cefis, E. Bartoloni, and M. Bonati, “Show me how to live: Firms’ financial conditions and innovation during the crisis,” Struct. Chang. Econ. Dyn., vol. 52, pp. 63–81, 2020. 2020 IEEE International Conference on Engineering, Technology and Innovation (ICE/ITMC) CEWP 20-10 The Short-Term Effect of COVID-19 on SelfEmployed Workers in Canada Louis-Philippe Béland Oluwatobi Fakorede Derek Mikola Carleton University Carleton University Carleton University June 2020 CARLETON ECONOMICS WORKING PAPERS Department of Economics 1125 Colonel By Drive Ottawa, Ontario, Canada K1S 5B6 The Short-Term Effect of COVID-19 on Self-Employed Workers in Canada * Louis-Philippe Beland Oluwatobi Fakorede Derek Mikola June 26th, 2020 Abstract Using the Canadian Labour Force Survey, we document the short-term impact of COVID-19 on self-employed individuals in Canada, which we interpret as small business owners. We document an important decrease in business ownership between February 2020 and May 2020 (-14.8 percent for incorporated and -10.1 percent for unincorporated entities). We find a greater decrease in ownership and aggregate hours for women, immigrants and less educated over the same period. The industries with the largest decrease are in art, culture, and recreation (-14.8 percent); in education, law and social, community and government services (-13.6 percent); and in sales and service occupations (-12.8 percent). Keywords: COVID-19, Self-Employed workers, Entrepreneurship, Employment, Labour Force, Hours. JEL codes: L26, J21, J24, I18 * Authors: Beland: Carleton University. Email: louisphilippe.beland@carleton.ca; Fakorede: Carleton University: Email: Olubunmifakorede@cmail.carleton.ca; Mikola: Carleton University. Email: Derekmikola@cmail.carleton.ca. We would like to thank Abel Brodeur for comments. 1 1 Introduction The COVID-19 pandemic has led the Canadian provincial governments to shut down nonessential businesses and services across Canada and impose social and physical distancing policies. These policies have resulted in severe job loss for Canadians. The pandemic is having unparalleled economic consequences on the Canadian economy. In this paper, we study the effect of the COVID-19 pandemic on self-employed individuals in Canada, which we interpret as small business owners. The importance of entrepreneurial activity and, in particular, small business entrepreneurship, has been widely discussed during the COVID-19 pandemic in Canada. The viability of small business is an important concern for policy makers since some small businesses may never recover from COVID-19 (e.g. Bensadoun (2020)).1 Small business owners play a pivotal role in the Canadian economy and it is crucial to understand how the pandemic is affecting them. In this paper, we use the Canadian Labour Force Survey, a nationally representative survey, to document the short-term effect of COVID-19 on the number of active small businesses in Canada. We then investigate if there are heterogeneous changes based on owners’ characteristics such as gender, immigrant status, education and age. We show an important decrease in business ownership between February 2020 and May 2020 (-14.8 percent for incorporated and -10.1 percent for unincorporated entities). Over the same period, we find larger decreases in ownership for immigrants (-16.1 percent), women (12.9 percent), and less educated individuals (-17.8 percent). We also find a drastic decrease in aggregate hours worked for immigrants (-44.3 percent), women (-43.5 percent), and less educated individuals (-57.2 percent). The industries with the largest decrease are in art, culture and recreation (-14.8 percent); in education, law and social, community and government services (-13.6 percent); and in sales and service occupations (-12.8 percent). Our results build on the growing literature studying the impact of the pandemic on economic outcomes in Canada (e.g., Beland, Brodeur, Mikola and Wright (2020); Beland, Brodeur, Haddad and Mikola (2020); Lange and Warman (2020)) and other countries (e.g., Beland, Brodeur and Wright (2020); Rojas et al. (2020); Brodeur et al. (2020); Kahn et al. (2020) and Lewandowski (2020)). We contribute to the literature by documenting the labour market effects of COVID-19 on self-employed workers in Canada, who represent a cornerstone of the Canadian economy. The rest of the paper is as follows: Section 2 discusses background and self-employment in Canada, Section 3 discusses the data, Section 4 presents the results, and Section 5 concludes. 1 A list of different policies implemented by the Canadian federal government is available here: canada.ca/en/department-finance/economic-response-plan.html 2 2 Background and Self-Employment in Canada Self-employed workers are a vital part of the Canadian economy which all levels of government have continuously promoted through the provisions of loans, grants, and other support. Self-employed people vary from owners of large incorporated businesses to owners of small unincorporated businesses. Programs such as the Canada Small Business Financing Program, where the government shares loan risk with lenders, exist to make it easier for small businesses to obtain funding for start up or expansion. Several studies underline the fragility of small businesses such as research by Bartik et al. (2020) based on a survey of 5,800 small businesses located in the US. Bartik et al. (2020) document that the median firm, of those with expenses over $10,000, only has enough liquidity to continue to carry out business for up to two weeks. Large shocks to these firms, like mandatory shut-downs, represent large strains which challenge their existence in the economy. A recent poll by the Canadian Imperial Bank of Commerce (CIBC), found that 81 percent of Canadian small business owners say COVID-19 has negatively impacted their operations.2 As part of the government’s COVID-19 Economic Response Plan, the government of Canada proposed to: provide a 75 percent wage subsidy for up to 3 months; allow businesses including the self-employed to defer GST/HST payments until June; provide a government guaranteed and funded loan to small businesses through the Canada Emergency Business Account; and launched a program that enables up to $45 billion in funding by further guaranteeing loans through Export Development Canada and the Business Development Bank. Several additional programs and policy have been created after COVID-19 such as business rent assistance.3 The attention given to small businesses is a result of how important they are seen to be in terms of job creation. Studies examining the effect of COVID-19 on small businesses are scarce as of yet, but one such study conducted by Fairlie (2020) on small businesses in the US examined demographic and industry patterns, using the Current Population Survey. Fairlie (2020) found that the number of business owners operating in the United States dropped by 22 percent between February 2020 to April 2020. This is a considerable drop, especially when compared to the Great recession, where there was a 5 percent drop in the number of small business owners. In examining demographic trends, Fairlie (2020) found that the African American population experienced the largest drop in the number of business owners due to COVID-19, followed by the Latinx population which lost 41 percent and 32 percent of their small businesses, respectively. His research further revealed that male business ownership decreased by 23 percent while female business ownership decreased by 19 percent. Our paper uses the Labour Force survey 2 The poll is available here: http://cibc.mediaroom.com/2020-05-04-COVID-19-impact-felt-by-81-per-centof-Canadian-small-business-owners-CIBC-Poll 3 For a complete list, see canada.ca/en/department-finance/economic-response-plan.html. It is also worth noting than provinces also created different programs to help small business owners. 3 to perform a similar analysis for the Canadian economy.4 A major concern is the possibility that racial and gender inequality will be exacerbated by the COVID-19 pandemic. According to Statistics Canada’s 2018 supplemental Labour Force Survey, about 15 percent of the Canadian population is self-employed and 38 percent are women. The same supplement to the Labour Force Survey revealed that the top reason individuals gave for choosing self-employment was “independence, freedom, being one’s own boss.” The second, third, and fourth most common reasons were, respectively,: the nature of the respondent’s job, work-family balance, and flexible hours respectively. One important disparity in the reasons for choosing self-employment when broken down by gender was that women were more likely to cite work-family balance (15 percent vs. 5 percent of men), and flexible hours (11 percent vs. 7 percent of men) as their main reason for being self-employed. This suggests that having kids is an important component of why women become self-employed. Moreover, mandatory daycare and school closures might have important repercussions for these selfemployed workers. There is a body of work examining why people go into self-employment such as the work of Boden (1999), Dawson et al. (2014) and Leonard et al. (2017).5 The Canadian literature has also looked at entry and exit from self-employment, including Chan (2019), Finnie and Gray (2018), Galindo da Fonseca (2019), Liu et al. (2018) and Schuetze (2015). We contribute to this literature by studying the effect of COVID-19 on entrepreneurship. There are a few related papers studying the effects of COVID-19 in Canada. Lange and Warman (2020) document a rapid decline in vacancy postings following the pandemic. Importantly, they find a similar impact across Canada. Beland, Brodeur, Mikola and Wright (2020) document the short-term impact of COVID-19 on labour market outcomes in Canada (see also Koebel and Pohler (2020) and Lemieux et al. (2020)). They find that COVID-19 had drastic negative effects, especially on younger, not married, and less educated workers. They also investigate the impact by occupation using indices for (1) exposed to disease, (2) work in proximity to co-workers, (3) essential workers, and (4) can work remotely. They find that the impact of COVID-19 was significantly more severe for indices (1) and (2), and significantly less for indices (3) and (4).6 Our paper documents the number of active business owners in Canada and investigates heterogeneity based on characteristics. Our results can help provide guidance for policy recommendations and highlight the self-employed workers in most need of assistance. 4 While we do not have data on race in the LFS, we find that immigrants are significantly more affected than Canadian born. 5 A large literature study the determinants of entrepreneurship. These factors include family background and race (e.g., Hout et al. (2000); and Fairlie and Robb (2010)), immigration status (e.g., Hunt and GauthierLoiselle (2010)), financial constraints (e.g., Fairlie and Krashinsky (2012)), risk attitudes (e.g., Blanchflower and Oswald (1998); Skriabikova et al. (2014)) and economic conditions (e.g., Fairlie (2013)). See Beland and Unel (2019) for a complete literature review. 6 Using the Canadian Perspective Survey, they also show that reported mental health is significantly lower among the most affected workers. Beland, Brodeur, Mikola and Wright (2020) study graphically self-employed workers and distinguish between incorporated and unincorporated. We study here in more detail the effect of COVID-19 on self-employed based on characteristics. 4 3 Data 3.1 Labour Force Survey The Canadian Labour Force Survey (LFS) presents a picture of the current state of the Canadian labour market. It is collected by Statistics Canada on a monthly basis and includes respondent-level information on employment status, weekly hours worked, and labour force status.7 It also includes characteristics of respondents such as gender, age, education level, marital status and immigration status. It is a nationally representative survey. The population surveyed includes non-institutionalized individuals 15 years of age and older in all of Canada. The survey excludes full-time members of the Canadian Armed Forces and people living on reserves or other Aboriginal settlements. Data is collected from interviews of 56,000 households and approximately 100,000 individuals on a monthly basis. Statistics Canada carries out the LFS survey via telephone in English or French or through visits to households by a field interviewer. The survey is cross-sectional, and the respondents remain in the survey for six months. Due to COVID-19, none of the interviews were conducted face-to-face after March 2020. The LFS allows us to document the short-term effect of COVID-19 on the number of active small businesses in Canada. We use changes in business ownership between February 2020 and May 2020 and we restrict our observations to those which are classified as selfemployed at their main job. Similarly to Fairlie (2020), we largely interpret self-employed workers as small business owners.8 The LFS also allows us to answer how COVID-19 has affected the labour market outcomes of the remaining self-employed workers and to investigate potential heterogeneity of the effect of COVID-19 on self-employed individuals. Our final dataset to create figures on the labour market outcomes of active self-employed workers uses the information of respondents collected between January 2016 and May 2020. Table 1 provides a description of our dataset. The top panel describes the main outcomes we use to summarize the labour market conditions of active self-employed workers in Canada. We see that the self-employed have a high likelihood of being in the labour force (97 percent, on average), a low likelihood of being unemployed (1 percent, on average), and work about 35.1 hours per week. The unemployed sample is restricted to the self-employed in the labour force while the labour force participation is the whole sample of self-employed. The total actual hours of work is restricted to the self-employed in the labour force. We focus on self-employed individuals between the ages of 15 and 69. The bottom panel shows the various demographic characteristics of the self-employed for the sample associated with the labour force participation market outcome from the top panel. 7 The public files of the LFS do not have information on hourly wages or earnings for self-employed. This information is available for individuals working in the public or private sector. 8 The literature argues that incorporated entities is a better proxy for entrepreneurship (e.g., Levine and Rubinstein (2017); Beland and Unel (2019)) 5 The LFS categorizes respondents into one of 4 different categories for the self employed which we collapse into incorporated and unincorporated.9 Table 1 shows that more than 60 percent of the self-employed are male. Among self-employed women, nearly 44 percent have children. Moreover, among the self-employed, 69 percent have a post-secondary accreditation and 70 percent are non-immigrants. 4 Results 4.1 Analysis of the number of active business owners before and after COVID-19 We investigate the effect of COVID-19 on the number of active self-employed business owners in Canada, following Fairlie (2020). Table 2 shows changes in relevant variables from January 2020 to May 2020. We interpret these as the number of active small businesses in the representative sample.10 The first column in the table shows that the number of unincorporated businesses decreased by 10.12 percent in the period between February and May 2020. The number of incorporated businesses decreased by a larger amount of 14.8 percent. Table 2 also shows the evolution of the number of small business owners employed over time as the consequences of COVID-19 became more dire. The last two columns in the first panel measure the total number of self-employed individuals that are employed with paid help and without paid help. We see a decrease in the number of business owners who reported being employed with paid help by approximately 10.1 percent between February and May 2020 and by 17.8 percent for those employed without paid help. To investigate how much less self-employed workers are working, we show the number of full-time as well as part-time aggregated hours worked at the main job. We find a decrease in aggregate hours for full-time self-employed workers by 34.4 percent and by 18.7 percent for self-employed workers usually working part-time. Furthermore, we see a decrease in the number of business owners that are in the labour force by 12.6 percent between February and May 2020. One potential concern over our analysis is related to the fact that the number of active business owners might have a seasonality component. To investigate this, we also report in our tables the changes from February 2019 to May 2019 and we compute a double-difference. Table 2 shows an increase in business owners between February 2019 and May 2019 and therefore all our double-difference estimates are significantly higher. We find that our calculations above, using changes from February to May 2020, could be a lower-bound of the effect of COVID-19 on self-employed business owners. Table 3 studies the number of business owners by characteristics of owners (gender, immigrant status, marital status, kids or no kids, education and age). Table 3 shows a larger decrease in ownership for female (-12.9 percent) versus male (-11.96 percent) owners from February 2020 to May 2020. It finds a higher decrease in ownership for immigrants (-16.1 percent) than non9 We aggregate “incorporated, with paid help” and “incorporated, no paid help” into “incorporated.” We make a similar aggregation for the unincorporated. 10 It should be noted that the actual number of small businesses is much greater, and the LFS surveys a small representative set of the population. We use the sample weight. 6 immigrants (-10.6 percent), and a higher decrease for not married (-15.4 percent) individuals compared to married (-10.5 percent) individuals over the same period. Table 3 finds a smaller decrease for women with kids (-9.2 percent) than without kids (-15.5 percent). This suggests that the daycare and schools being closed might not be the primary reason for the decrease in business ownership by women. Table 3 also presents results by education category and age category. It shows a higher decrease for less educated business owners (-17.8 percent for individuals with less than a high school diploma) and individuals in the 35 to 54 years age group (-13.8 percent). Table 3 also presents results for double-difference calculation, using change from February to May 2019. Once again, the changes from February to May 2019 were all positive and all our double-difference estimates confirm the previous analyses but show impacts which are substantially larger in magnitude. Table 4 presents the aggregate hours worked by individual characteristics (gender, immigration status, marital status, kids or no kids, education and age) before and after COVID-19. Table 4 finds a drastic decrease in aggregate hours for all characteristics. The effects are significantly larger for immigrants (-44.3 percent), women (-43.5 percent), individuals with less than high school education (-57.2 percent), and younger workers (-49.5 percent for aged 15 to 34 and -44.9 percent for aged 35 to 54) from February 2020 to May 2020. We also present a double-difference calculation and the effect of COVID-19 are significantly larger using this method. Moreover, the double-difference calculation do not alter which self-employed workers were more affected by COVID-19.11 Table 5 studies the effect of COVID-19 on the number of active businesses by industry. It compares the first five months of 2020 to the first five months of 2019. It shows a reduction in the number of small businesses for most industries. One important exception is those in health. There was an increase in small businesses in health by 11.4 percent for the first five months of 2020 compared to the first five months of 2019. The industries with the largest decrease are in art, culture and recreation (-14.8 percent); in education, law and social, community, and government services (-13.6 percent); and in sales and service occupations (-12.8 percent). 4.2 Graphical Analysis of the labour market outcomes of the active self-employed business owners before and after COVID-19 Next, we present graphical evidences to study changes in labour market outcomes for the active self-employed workers from January 2016 to May 2020. The LFS gives us critical labour market information about the self-employed which includes whether they are employed, in the labour force and their total actual hours of work.12 We construct our figures through weighted 11 Similarly to Beland, Brodeur, Mikola and Wright (2020) which studies the effect of COVID-19 on the Canadian labour market, we find a more pronounced impact on younger and less-educated self-employed workers. The larger impact on immigrants and women seems to be self-employed specific. 12 As mentioned above, the public data files of the LFS do not have information on hourly wages or earnings for the self-employed, precluding analysis on changes in wage rates over time. It is possible that COVID-19 affects earnings differently depending on incorporation status. 7 aggregation of the self-employed by year, month and various demographic characteristics. As discussed earlier, for each labour market outcome, we restrict our dataset as follows: the unemployed sample is restricted to the self-employed in the labour force. The total actual hours of work are restricted to the self-employed in the labour force. We focus on self-employed individuals between the ages of 15 and 69. Our analysis can help highlight the active selfemployed in most need of help. All graphs are structured in the same way. Panel (a) presents results for unemployment rate, Panel (b) for labour force participation, and Panel (c) for actual hours of work over the period January 2016 to May 2020.13 Figure 1 shows that COVID-19 led to an increase in unemployment (from about 0.6 to 1.4 percent from February 2020 to May 2020), a decrease in labour force participation (from about 96.5 to 93.5 percent between February 2020 to May 2020), and a drastic decrease in actual hours worked (from about 34 to 23 hours from February 2020 to May 2020) for self-employed workers that remained in the labour force. Figure 2 distinguishes between self-employed incorporated and unincorporated. Self-employed are separated into two categories: incorporated (working for themselves in corporate entities) and unincorporated (working for themselves in other entities). Figure 2 shows that both incorporated and unincorporated are both negatively affected by COVID-19, suggesting that COVID-19 has had a negative impact on entrepreneurship. Figure 3 distinguishes between self-employed business owners with paid held and without paid help. Figure 3 shows that the effect of COVID-19 is significantly larger (for unemployment, labour force participation and actual hours worked) for self-employed without paid help. This could be due to their businesses being smaller operations. We now investigate with graphical representations the short-term effects of COVID-19 on labour market outcomes for different subgroups of respondents. Different groups of selfemployed workers are potentially affected differently and this could help target any additional policy aimed at helping small business owners. Figures 4, 5, 6, 7, 8, and 9, illustrate the outcome variables by gender, age group, marital status, education, immigration status, and years since immigration, respectively. Figure 4 shows our three main outcome variables by gender. It shows that COVID-19 resulted in negative labour market outcomes for both male and female self-employed workers, but the effect is significantly larger for women, especially for unemployment. Figure 5 shows the evolution of the three outcome variables by age group. It shows that COVID-19 was particularly difficult for younger self-employed, possibly due to their business being less established. Figure 6 presents the impact of COVID-19 by marital status. Figure 6 shows COVID-19 has had a negative impact for both married and unmarried self-employed individuals. However, the effect of COVID-19 is larger for unmarried self-employed workers. Next, Figure 7 shows the effect of COVID-19 by education. It finds that COVID-19 had 13 Data on immigration status is available in the LFS only after December 2016, which leads to a slightly shorter time period of between January 2017 and May 2020. 8 a negative impact on all groups. In Figure 8 we see a negative impact of COVID-19 for both immigrants and non-immigrants, but the effect is more pronounced for immigrants. Similarly, Figure 9 shows that the effect is more pronounced for immigrants of less than 10 years, also potentially due to their business being less established. 5 Conclusion In this paper, we study the effect of the COVID-19 pandemic on labour market outcomes of self-employed workers in Canada. The importance of small business entrepreneurship has been widely discussed in Canada during the COVID-19 pandemic. The viability of small business is an important concern for policy makers and there are concerns that some small businesses may never recover (e.g. Bensadoun (2020)). We use the Labour Force Survey, a nationally representative survey, to document the shortterm effect of COVID-19 on the number of active small businesses in Canada. We also investigate if there are heterogeneous changes based on owners’ characteristics such as gender, immigrant status, education, and age. We show an important decrease in business ownership between February 2020 and May 2020 (-14.8 percent for incorporated and -10.1 percent for unincorporated entities). We find a higher decrease in ownership for immigrants (-16.1 percent), women (-12.9 percent), less educated individuals (17.8 percent) over the same period. We also find a drastic decrease in aggregate hours worked for immigrants (-44.3 percent), women (-43.5 percent), and less educated individuals (-57.2 percent). The industries with the largest decrease are in art, culture and recreation (-14.8 percent); in education, law and social, community and government services (-13.6 percent); and in sales and service occupations (-12.8 percent). The Federal government has been responding to fear of long-term negative effects on businesses using several programs. Despite this help, we document considerable negative short-term effects on small businesses. We also document several heterogeneity based on owners characteristics which can help target additional help. Future research needs to follow closely what happens to small businesses and investigate if the shutdown of businesses is short-term or permanent. This can potentially have a long term impact on the Canadian economy and on job creation. References Bartik, A. W., Bertrand, M., Cullen, Z. B., Glaeser, E. L., Luca, M. and Stanton, C. T.: 2020, How are small businesses adjusting to covid-19? early evidence from a survey, Working Paper 26989, National Bureau of Economic Research. Beland, L.-P., Brodeur, A., Haddad, J. and Mikola, D.: 2020, Covid-19, family stress and domestic violence: Remote work, isolation and bargaining power, IZA Discussion Paper . 9 Beland, L.-P., Brodeur, A., Mikola, D. and Wright, T.: 2020, The short-term economic consequences of covid-19: Occupation tasks and mental health in canada. IZA Discussion Paper. Beland, L.-P., Brodeur, A. and Wright, T.: 2020, Covid-19, stay-at-home orders and employment: Evidence from cps data, IZA Discussion Paper . Beland, L.-P. and Unel, B.: 2019, Politics and Entrepreneurship in the US, Canadian Journal of Economics/Revue Canadienne D’e?conomique 52(1), 33–57. Bensadoun, E.: 2020, 1 in 10 canadian businesses may never recover from coronavirus, industry group says, Global News . Blanchflower, D. G. and Oswald, A. J.: 1998, What makes an entrepreneur?, Journal of labor Economics 16(1), 26–60. Boden, R. J.: 1999, Flexible Working Hours, Family Responsibilities, and Female SelfEmployment: Gender Differences in Self-Employment Selection, The American Journal of Economics and Sociology 58(1), 71–83. Brodeur, A., Clark, A. E., Fle?che, S. and Powdthavee, N.: 2020, COVID-19, Lockdowns and Well-Being: Evidence from Google Trends. IZA Discussion Paper 13204. Chan, J.: 2019, Tariffs and the composition of employment: Evidence from the canada–us free trade agreement, Canadian Public Policy 45(3), 342–365. Dawson, C., Andrew, H. and Latreille, P.: 2014, Individual Motives for Choosing Selfemployment in the UK: Does Region Matter?, Regional Studies 48(5), 804–822. Fairlie, R. W.: 2013, Entrepreneurship, economic conditions, and the great recession, Journal of Economics & Management Strategy 22(2), 207–231. Fairlie, R. W.: 2020, The Impact of COVID-19 on Small Business Owners: Evidence of EarlyStage Losses from the April 2020 Current Population Survey. IZA Discussion Paper 13311. Fairlie, R. W. and Krashinsky, H. A.: 2012, Liquidity constraints, household wealth, and entrepreneurship revisited, Review of Income and Wealth 58(2), 279–306. Fairlie, R. W. and Robb, A. M.: 2010, Race and entrepreneurial success: Black-, Asian-, and White-owned businesses in the United States, MIT Press. Finnie, R. and Gray, D.: 2018, How do older laid-off workers get by: Reemployment, early retirement, or social insurance benefits?, Canadian Public Policy 44(2), 173–189. Galindo da Fonseca, J.: 2019, Unemployment, entrepreneurship and firm outcomes. Hout, M., Rosen, H. et al.: 2000, Self-employment, family background, and race, Journal of Human Resources 35(4), 670–692. Hunt, J. and Gauthier-Loiselle, M.: 2010, How much does immigration boost innovation?, American Economic Journal: Macroeconomics 2(2), 31–56. Kahn, L. B., Lange, F. and Wiczer, D. G.: 2020, Labor demand in the time of covid-19: Evidence from vacancy postings and ui claims, Technical report, National Bureau of Economic Research. 10 Koebel, K. and Pohler, D.: 2020, Labor markets in crisis: The causal impact of canada’s covd19 economic shutdown on hours worked for workers across the earnings distribution, Technical report, Working Paper Series. Lange, F. and Warman, C.: 2020, Vacancy posting in 2020: Estimates based on job bank and external providers. Lemieux, T., Milligan, K., Schirle, T. and Skuterud, M.: 2020, Initial impacts of the covid-19 pandemic on the canadian labour market, Technical report, Working Paper Series. Leonard, P. S., Emery, J. H. and McDonald, J. T.: 2017, Push or pull into self employment? evidence from longitudinal canadian tax data, working paper . Levine, R. and Rubinstein, Y.: 2017, Smart and Illicit: Who Becomes an Entrepreneur and Do they Earn More?, Quarterly Journal of Economics 132(2), 963–1018. Lewandowski, P.: 2020, Occupational exposure to contagion and the spread of covid-19 in Europe. IZA Discussion Paper 13227. Liu, H., Grekou, D. et al.: 2018, The entry into and exit out of self-employment and business ownership in canada, Statistics Canada, Analytical Studies Branch . Rojas, F. L., Jiang, X., Montenovo, L., Simon, K. I., Weinberg, B. A. and Wing, C.: 2020, Is the cure worse than the problem itself? immediate labor market effects of covid-19 case rates and school closures in the us, Technical report, National Bureau of Economic Research. Schuetze, H. J.: 2015, Self-employment and retirement in canada: The labour force dynamics of older workers, Canadian Public Policy 41(1), 65–85. Skriabikova, O. J., Dohmen, T. and Kriechel, B.: 2014, New evidence on the relationship between risk attitudes and self-employment, Labour Economics 30, 176–184. 11 Table 1: Summary Statistics for Labour Market and Individual Variables Labour Market Variables Mean Std. Dev. Max Min Unemployed 0.01 0.08 1.00 0.00 Labour Force Participation 0.97 0.18 1.00 0.00 Total Actual Hours of Work 35.11 22.33 99.00 0.00 Individual Variables Labour Force Participation Specification (%) Self Employment Incorporation Status Unincorporated Incorporated 55.0 45.0 Sex Male Female 62.7 37.3 Age Categories 15 to 34 35 to 54 55+ 17.2 46.3 36.5 Highest Educational Attainment Less than highschool Highschool or some college Postsecondary Accreditation 8.8 22.5 68.7 Women with Kids Woman, no kids Woman with Kids 56.1 43.9 Immigration Immigration ≤ 10 years ago Immigration > 10 years ago Non-immigrant 6.4 23.5 70.1 Marital Status Not Married Married 38.1 61.9 Notes: Authors’ calculations using the Labour Force Survey between January 2016 and May 2020. All observations are self employed. 12 Table 2: Active Small Business Statistics Before and After COVID-19 Paid Help Unincorporated Incorporated With Without Jan 2020 Feb 2020 Mar 2020 Apr 2020 May 2020 4,768 4,693 4,417 4,300 4,218 4,115 4,102 3,761 3,469 3,495 6,318 6,254 5,925 5,654 5,622 2,565 2,540 2,253 2,115 2,091 Feb 2019 May 2019 4,775 5,009 4,107 4,142 6,370 6,427 2,512 2,724 May - Feb 2020 (?%) May - Feb 2019 (?%) -10.12% 4.92% -14.80% 0.83% -10.11% 0.90% -17.69% 8.42% 2020 - 2019 (?%) -15.04% -15.63% -11.01% -26.12% Hours worked ≥ 30 (Millions) Hours worked < 30 (Millions) In labour force Jan 2020 Feb 2020 Mar 2020 Apr 2020 May 2020 77.2 78.2 53.8 42.5 51.3 11.1 11.0 10.0 7.8 8.9 8,931 8,822 8,211 7,793 7,712 Feb 2019 May 2019 75.0 85.8 11.0 11.4 8,945 9,216 May - Feb 2020 (?%) May - Feb 2019 (?%) -34.38% 14.40% -18.71% 3.36% -12.59% 3.02% 2020 - 2019 (?%) -48.77% -22.07% -15.61% Notes: Authors’ calculations. Data from the Canadian Labour Force Survey. The time period is January 2020 to May 2020. Observations are only those who are self-employed. Weights are applied to all measures. Counts are weighted within the sample while hours worked are weighted up to the population. 13 Table 3: Business Owners by Individual Characteristics Before and After COVID-19 (a) Business Owners by Individual Characteristics Male Female Not Immigrant Immigrant Jan 2020 Feb 2020 Mar 2020 Apr 2020 May 2020 5638 5492 5092 4866 4835 3245 3303 3086 2903 2878 6085 6028 5762 5387 5391 2798 2767 2416 2382 2322 Feb 2019 May 2019 5615 5714 3267 3437 6222 6480 2660 2671 May - Feb 2020 (?%) May - Feb 2019 (?%) -11.97% 1.76% -12.86% 5.21% -10.56% 4.16% -16.09% 0.39% 2020 - 2019 (?%) -13.72% -18.07% -14.72% -16.48% Not Married Married No kids With kids Jan 2020 Feb 2020 Mar 2020 Apr 2020 May 2020 3362 3255 3090 2889 2755 5521 5540 5088 4880 4958 1864 1855 1700 1610 1568 1443 1489 1429 1346 1353 Feb 2019 May 2019 3405 3609 5477 5542 1810 1887 1431 1534 May - Feb 2020 (?%) May - Feb 2019 (?%) -15.37% 6.00% -10.50% 1.18% -15.45% 4.25% -9.16% 7.20% 2020 - 2019 (?%) -21.37% -11.68% -19.71% -16.36% Women Notes: Authors’ calculations. Data from the Canadian Labour Force Survey with within-sample weights applied. The time period is January 2020 to May 2020. Observations are only those who are self-employed. 14 (b) Business Owners by Individual Characteristics Education Level Less than high school High school or some college Postsecondary accreditation Jan 2020 Feb 2020 Mar 2020 Apr 2020 May 2020 735 698 630 595 574 1883 1855 1774 1708 1650 6266 6242 5774 5466 5489 Feb 2019 May 2019 710 744 1947 2070 6225 6337 May - Feb 2020 (?%) May - Feb 2019 (?%) -17.81% 4.70% -11.03% 6.34% -12.06% 1.80% 2020 - 2019 (?%) -22.51% -17.38% -13.86% Age Categories 15 to 34 35 to 54 55+ Jan 2020 Feb 2020 Mar 2020 Apr 2020 May 2020 1459 1414 1357 1338 1313 4153 4142 3857 3588 3571 3271 3240 2964 2843 2829 Feb 2019 May 2019 1511 1540 4112 4331 3259 3280 May - Feb 2020 (?%) May - Feb 2019 (?%) -7.17% 1.90% -13.77% 5.34% -12.67% 0.64% 2020 - 2019 (?%) -9.07% -19.10% -13.31% Notes: Authors’ calculations. Data from the Canadian Labour Force Survey, with within-sample weights applied. The time period is January 2020 to May 2020. Observations are only those who are self-employed. 15 Table 4: Aggregate Hours Worked by Individual Characteristics Before and After COVID-19 (a) Aggregate Hours Worked by Individual Characteristics Male Female Not Immigrant Immigrant Jan 2020 Feb 2020 Mar 2020 Apr 2020 May 2020 61.4 60.7 46.4 36.8 44.2 26.9 28.5 17.3 13.5 16.1 59.7 60.9 46.1 36.9 44.5 28.6 28.3 17.7 13.4 15.8 Feb 2019 May 2019 59.3 66.3 26.7 30.9 59.0 69.1 27.0 28.0 May - Feb 2020 (?%) May - Feb 2019 (?%) -27.28% 11.76% -43.46% 15.71% -26.95% 17.17% -44.26% 3.82% 2020 - 2019 (?%) -39.04% -59.16% -44.12% -48.08% Women Not Married Married No kids With kids Jan 2020 Feb 2020 Mar 2020 Apr 2020 May 2020 32.5 32.9 22.8 16.9 20.4 55.8 56.3 41.0 33.4 39.9 13.9 14.5 9.2 6.9 8.1 13.0 14.0 8.0 6.6 8.0 Feb 2019 May 2019 32.9 38.1 53.1 59.0 14.0 15.6 12.7 15.2 May - Feb 2020 (?%) May - Feb 2019 (?%) -38.16% 15.91% -29.10% 11.17% -44.06% 11.41% -42.82% 20.47% 2020 - 2019 (?%) -54.07% -40.27% -55.48% -63.29% Notes: Authors’ calculations. Data from the Canadian Labour Force Survey, weighted up to the population. The time period is January 2020 to May 2020. All values are in millions of hours. Observations are only those who are self-employed. 16 (b) Aggregate Hours Worked by Individual Characteristics Education Level Less than high school High school or some college Postsecondary accreditation Jan 2020 Feb 2020 Mar 2020 Apr 2020 May 2020 1.2 1.5 0.8 0.7 0.6 4.6 5.0 3.1 2.3 2.7 21.0 22.0 13.4 10.5 12.7 Feb 2019 May 2019 6.7 7.7 18.5 22.8 60.8 66.6 May - Feb 2020 (?%) May - Feb 2019 (?%) -57.21% 15.65% -45.66% 23.52% -42.03% 9.49% 2020 - 2019 (?%) -72.85% -69.18% -51.52% Age Categories 15 to 34 35 to 54 55+ Jan 2020 Feb 2020 Mar 2020 Apr 2020 May 2020 4.5 5.0 2.8 2.1 2.5 14.6 15.6 9.3 7.3 8.6 7.8 7.8 5.3 4.1 5.0 Feb 2019 May 2019 15.1 17.5 45.1 50.6 25.7 29.0 May - Feb 2020 (?%) May - Feb 2019 (?%) -49.51% 15.75% -44.88% 12.09% -36.70% 12.93% 2020 - 2019 (?%) -65.26% -56.97% -49.63% Notes: Authors’ calculations. Data from the Canadian Labour Force Survey, weighted up to the population. The time period is January 2020 to May 2020. All values are in millions of hours. Observations are only those who are self-employed. 17 Table 5: Small Business Statistics by Industry January to May 2018 2019 2020 2020 - 2019 (?%) Business, finance and administration occupations 4,640 4,601 4,065 -11.65% Health occupations 3,411 3,405 3,794 11.43% Management occupations 11,118 10,532 9,888 -6.12% Natural and applied sciences and related occupations 2,689 2,804 2,597 -7.36% Natural resources, agriculture and related occupations 1,088 962 1,007 4.62% Occupations in art, culture, recreation and sport 3,559 3,592 3,059 -14.84% Occupations in education, law and social, community and government services 3,646 3,578 3,090 -13.64% Occupations in manufacturing and utilities 732 688 702 Sales and service occupations 7,095 7,178 6,260 -12.79% Trades, transport and equipment operators and related occupations 7,503 7,649 6,876 -10.11% Total 45,482 44,989 41,338 -8.12% 2.02% Notes: Authors’ calculations. Data from the Canadian Labour Force Survey with within-sample weights applied. Observations are only those who are self-employed. 18 Figure 1: Unemployment Rate, Labour Force Participation, and Hours of Work for SelfEmployed. (a) Unemployment Rate .4 2016m1 97 96 95 94 .6 Unemployment Rate (%) .8 1 1.2 Labour Force Participation Rate (%) 1.4 98 (b) Labour Force Participation 2017m1 2018m1 2019m1 2020m1 2016m1 2017m1 2018m1 2019m1 2020m1 20 Total Actual Weekly Hours Worked 25 30 35 40 (c) Hours of Work 2016m1 2017m1 2018m1 2019m1 2020m1 Notes: Authors’ calculations. Data from the Canadian Labour Force Survey. The time period is January 2016 to May 2020. 19 Figure 2: Unemployment Rate, Labour Force Participation, and Hours of Work for SelfEmployed by Incorporation Status. (a) Unemployment Rate 0 2016m1 98 96 94 92 Unemployment Rate (%) .5 1 1.5 Labour Force Participation Rate (%) 2 100 (b) Labour Force Participation 2017m1 2018m1 2019m1 Unincorporated 2020m1 2016m1 Incorporated 2017m1 2018m1 Unincorporated 2019m1 2020m1 Incorporated 15 Total Actual Weekly Hours Worked 20 25 30 35 40 (c) Hours of Work 2016m1 2017m1 2018m1 Unincorporated 2019m1 2020m1 Incorporated Notes: Authors’ calculations. Data from the Canadian Labour Force Survey. The time period is January 2016 to May 2020. 20 Figure 3: Unemployment Rate, Labour Force Participation, and Hours of Work for SelfEmployed by those with Paid Help (a) Unemployment Rate 0 2016m1 98 96 94 92 Unemployment Rate (%) .5 1 1.5 Labour Force Participation Rate (%) 2 100 (b) Labour Force Participation 2017m1 2018m1 2019m1 No paid help 2020m1 2016m1 Has paid help 2017m1 2018m1 No paid help 2019m1 2020m1 Has paid help 10 Total Actual Weekly Hours Worked 20 30 40 50 (c) Hours of Work 2016m1 2017m1 2018m1 No paid help 2019m1 2020m1 Has paid help Notes: Authors’ calculations. Data from the Canadian Labour Force Survey. The time period is January 2016 to May 2020. 21 0 2016m1 97 96 95 94 93 Unemployment Rate (%) .5 1 1.5 Labour Force Participation Rate (%) 2 98 Figure 4: Unemployment Rate, Labour Force Participation, and Hours of Work for SelfEmployed by Sex. 2017m1 2018m1 2019m1 Male 2020m1 2016m1 Female 2017m1 2018m1 Male 2020m1 Female (b) Labour Force Participation. 15 Total Actual Weekly Hours Worked 20 25 30 35 40 (a) Unemployment Rate. 2019m1 2016m1 2017m1 2018m1 Male 2019m1 2020m1 Female (c) Hours of Work. Notes: Authors’ calculations. Data from the Canadian Labour Force Survey. The time period is January 2016 to April 2020. Observations are only those who are self-employed. 22 0 2016m1 98 96 94 92 Unemployment Rate (%) 1 2 Labour Force Participation Rate (%) 3 100 Figure 5: Unemployment Rate, Labour Force Participation, and Hours of Work for SelfEmployed by Age Group. 2017m1 15 to 34 2018m1 2019m1 35 to 54 2020m1 2016m1 55+ 2017m1 15 to 34 2019m1 35 to 54 2020m1 55+ (b) Labour Force Participation. 15 Total Actual Weekly Hours Worked 20 25 30 35 40 (a) Unemployment Rate. 2018m1 2016m1 2017m1 2018m1 15 to 34 2019m1 35 to 54 2020m1 55+ (c) Hours of Work. Notes: Authors’ calculations. Data from the Canadian Labour Force Survey. The time period is January 2016 to April 2020. Observations are only those who are self-employed. 23 0 2016m1 96 94 92 Unemployment Rate (%) .5 1 1.5 Labour Force Participation Rate (%) 2 98 Figure 6: Unemployment Rate, Labour Force Participation, and Hours of Work for SelfEmployed by Marital Status. 2017m1 2018m1 2019m1 Not Married 2020m1 2016m1 Married 2017m1 2018m1 Not Married 2020m1 Married (b) Labour Force Participation. 15 Total Actual Weekly Hours Worked 20 25 30 35 40 (a) Unemployment Rate. 2019m1 2016m1 2017m1 2018m1 Not Married 2019m1 2020m1 Married (c) Hours of Work. Notes: Authors’ calculations. Data from the Canadian Labour Force Survey. The time period is January 2016 to April 2020. Observations are only those who are self-employed. 24 0 2016m1 96 94 92 90 Unemployment Rate (%) .5 1 1.5 Labour Force Participation Rate (%) 2 98 Figure 7: Unemployment Rate, Labour Force Participation, and Hours of Work for SelfEmployed by Education. 2017m1 Less than high school 2018m1 2019m1 High school or some college 2020m1 2016m1 Postsecondary Accreditation 2017m1 Less than high school 2019m1 High school or some college 2020m1 Postsecondary Accreditation (b) Labour Force Participation. 20 Total Actual Weekly Hours Worked 25 30 35 40 (a) Unemployment Rate. 2018m1 2016m1 2017m1 Less than high school 2018m1 2019m1 High School or some college 2020m1 Postsecondary Accreditation (c) Hours of Work. Notes: Authors’ calculations. Data from the Canadian Labour Force Survey. The time period is January 2016 to April 2020. Observations are only those who are self-employed. 25 0 2017m1 98 96 94 92 .5 Unemployment Rate (%) 1 1.5 2 Labour Force Participation Rate (%) 2.5 100 Figure 8: Unemployment Rate, Labour Force Participation, and Hours of Work for SelfEmployed by Immigration Status. 2018m1 2019m1 Not Immigrant 2020m1 2017m1 2018m1 Immigrant Not Immigrant 2020m1 Immigrant (b) Labour Force Participation. 15 Total Actual Weekly Hours Worked 20 25 30 35 40 (a) Unemployment Rate. 2019m1 2017m1 2018m1 2019m1 Not Immigrant 2020m1 Immigrant (c) Hours of Work. Notes: Authors’ calculations. Data from the Canadian Labour Force Survey. The time period is January 2017 to April 2020. Observations are only those who are self-employed. 26 0 2017m1 98 96 94 92 1 Unemployment Rate (%) 2 3 4 Labour Force Participation Rate (%) 5 100 Figure 9: Unemployment Rate, Labour Force Participation, and Hours of Work for SelfEmployed by Years Since Immigration. 2018m1 Immigration, Less than 10 years ago 2019m1 Immigration, Greater than 10 years ago 2020m1 2017m1 2018m1 Immigration, Less than 10 years ago Not Immigrant 2020m1 Not Immigrant (b) Labour Force Participation. 15 Total Actual Weekly Hours Worked 20 25 30 35 40 (a) Unemployment Rate. 2019m1 Immigration, Greater than 10 years ago 2017m1 2018m1 2019m1 Immigration, Less than 10 years ago Immigration, Greater than 10 years ago 2020m1 Not Immigrant (c) Hours of Work. Notes: Authors’ calculations. Data from the Canadian Labour Force Survey. The time period is January 2017 to April 2020. Observations are only those who are self-employed. 27 Annu. Rev. Psychol. 1999. 50:537–67 Copyright ã 1999 by Annual Reviews. All rights reserved SURVEY RESEARCH Jon A. Krosnick Department of Psychology, Ohio State University, Columbus, Ohio 43210; e-mail: krosnick@osu.edu KEY WORDS: surveys, interviewing, polls, questionnaires, pretesting ABSTRACT For the first time in decades, conventional wisdom about survey methodology is being challenged on many fronts. The insights gained can not only help psychologists do their research better but also provide useful insights into the basics of social interaction and cognition. This chapter reviews some of the many recent advances in the literature, including the following: New findings challenge a long-standing prejudice against studies with low response rates; innovative techniques for pretesting questionnaires offer opportunities for improving measurement validity; surprising effects of the verbal labels put on rating scale points have been identified, suggesting optimal approaches to scale labeling; respondents interpret questions on the basis of the norms of everyday conversation, so violations of those conventions introduce error; some measurement error thought to have been attributable to social desirability response bias now appears to be due to other factors instead, thus encouraging different approaches to fixing such problems; and a new theory of satisficing in questionnaire responding offers parsimonious explanations for a range of response patterns long recognized by psychologists and survey researchers but previously not well understood. CONTENTS INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SAMPLING AND RESPONSE RATES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PRETESTING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RIGID INTERVIEWING VERSUS CONVERSATIONAL INTERVIEWING . . . . . . . . . . QUESTIONNAIRE DESIGN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Open versus Closed Questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Labeling of Rating-Scale Points. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Role of Conversational Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0084-6570/99/0201-0537$08.00 538 539 541 542 543 543 544 545 537 538 KROSNICK Social Desirability Bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Optimizing versus Satisficing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 545 546 559 560 INTRODUCTION These are exciting times for survey research. The literature is bursting with new insights that demand dramatic revisions in the conventional wisdom that has guided this research method for decades. Such dramatic revisions are nothing new for survey researchers, who are quite experienced with being startled by an unexpected turn of events that required changing their standard practice. Perhaps the best known such instance involved surveys predicting US election outcomes, which had done reasonably well at the start of the twentieth century (Robinson 1932). But in 1948 the polls predicted a Dewey victory in the race for the American presidency, whereas Truman actually won easily (Mosteller et al 1949). At fault were the nonsystematic methods used to generate samples of respondents, so we learned that representative sampling methods are essential to permit confident generalization of results. Such sampling methods soon came into widespread use, and survey researchers settled into a “standard practice” that has stood relatively unchallenged until recently (for lengthy discussions of the method, see Babbie 1990; Lavrakas 1993; Weisberg et al 1996). This standard practice included not only the notion that systematic, representative sampling methods must be used, but also that high response rates must be obtained and statistical weighting procedures must be imposed to maximize representativeness. Furthermore, although face-to-face interviewing was thought to be the optimal method, the practicalities of telephone interviewing made it the dominant mode since the mid-1980s. Self-administered mail surveys were clearly undesirable, because they typically obtained low response rates. And although a few general rules guided questionnaire design (e.g. Parten 1950), most researchers viewed it as more of an art than a science. There is no best way to design a question, said proponents of this view; although different phrasings or formats might yield different results, all are equally informative in providing insights into the minds of respondents. Today, this conventional wisdom is facing challenges from many directions. We have a refreshing opportunity to rethink how best to implement surveys and enhance the value of research findings generated using this method. This movement has three valuable implications for psychology. First, researchers who use the survey method to study psychological phenomena stand to benefit, because they can enhance the validity of their substantive results by using new methodologies, informed by recent lessons learned. Second, these SURVEY RESEARCH 539 insights provide opportunities to reconsider past studies, possibly leading to recognize that some apparent findings were illusions. Third, many recent lessons provide insights into the workings of the human mind and the unfolding of social interaction. Thus, these insights contribute directly to the building of basic psychological theory. Because recent insights are so voluminous, this chapter can describe only a few, leaving many important ones to be described in future Annual Review of Psychology chapters. One significant innovation has been the incorporation of experiments within surveys, thus permitting strong causal inference with data from representative samples. Readers may learn about this development from a chapter in the Annual Review of Sociology (Sniderman & Grob 1996). The other revelations, insights, and innovations discussed here are interesting because they involve the overturning of long-standing ideas or the resolution of mysteries that have stumped researchers for decades. They involve sampling and response rates, questionnaire pretesting, interviewing, and questionnaire design. SAMPLING AND RESPONSE RATES One hallmark of survey research is a concern with representative sampling. Scholars have, for many years, explored various methods for generating samples representative of populations, and the family of techniques referred to as probability sampling methods do so quite well (e.g. Henry 1990, Kish 1965). Many notable inac...