Unit 5 Reliability assignment   You’ll be using the dataset called math attitude questions

Unit 5 Reliability assignment

You’ll be using the dataset called math attitude questions.sav for this assignment. In this dataset, participants rated themselves on a scale from 1 to 4 where 1 = very atypical of me to 4 = very typical of me. The questions are listed below.

Item01: I practice math skills until I can do them well.

Item02: I feel happy after solving a hard problem.

Item03: I solve math problems quickly.

Item04: I give up easily instead of persisting if a math problem is difficult.

Item05: I am a little slow catching on to new problems in math.

Item06: I do not get much pleasure out of math problems.

Item07: I prefer to figure out how to solve problems without asking for help.

Item08: I do not keep at it very long when a math problem is challenging.

Item09: I am very competent in math.

Item10: I smile only a little (or not at all) when I solve a math problem.

Item11: I have some difficulties doing math as well as other kids my age.

Item12: I try to complete my math problems even if it takes a long time to finish.

Item13: I explore all possible solutions of a complex problem before going on to another one.

Item14: I really enjoy doing math problems.

1. First, read through the questions and find out four questions that were negatively worded. (5%)  

********Those four negatively worded questions have been recoded for you on the dataset. Be sure to use the recoded variables at the end of the data set when you run your analyses. Substitute the recoded variables for their corresponding item number. Don’t use BOTH!

2. Group items 1-14 into three different subscales based on the content of the items (Attitude, Emotion, Competency). (15%)
3. Then run an internal consistency reliability Cronbach’s alpha for EACH subscale that you determined. (40%)
4. Interpret the output and write your results for each subscale. You should list the items that go with each subscale. Discuss whether you want to delete, reword or change any items from any scale. (40%)

Depending on how you group your questions to make subscales, your analysis might be different from others. That’s OK.