In this situation, we are told that a teacher started a new reading program, and after 10 weeks, the average reading speed of second graders was measured at 92.8 wpm. What kind of results is this teacher allowed to report? Not many... There are many problems with the setup of this situation. Let's look at them more closely.

How long is the program supposed to last?

We are told that the student's reading speed was checked 10 weeks after the program was started. Is this the end of the program? Is this sometime in the middle of the program? Does the program have defined benchmarks or cycles? This time of 10 weeks seems arbitrary and should be clearly explained as to its significance.

Does this work only for 2nd graders?

We are told that they measured the reading speeds of a group of second graders. Was this who the program was targeted at? Should all students at this school be benefiting from this program? Do all students have access to the processes that the new program instituted? Were the second graders the only group that showed any kind of promise from using the program? The fact that only the second graders were reported on seems, again, arbitrary and raises more questions about the rest of the students at the school.

Does the new mean reading speed indicate a significant change?

We are told that the mean reading speed of these students is now 92.8 wpm. Is that an increase? Has this average decreased? Has it stayed the same (or close enough)? This result does not indicate any improvement in reading speed. If the initial reading speed (before the new program was instituted) was not measured, there is no value to be taken from this average speed taken after the program was run.

So, what might be concluded based on this result? All we know for sure is that the mean reading speed of the second graders at this school is now around 92.8 wpm. We cannot say anything about the effectiveness of the program. This scenario brings to light some of the reasons it is important to plan out an experiment like this before implementing it. A researcher could go through a lot of work - namely, creating a new reading program, administering it, and testing students on their reading speeds - and not be able to make any kind of meaningful conclusions after the fact.