Sample Response: Sample Paper #2

Research Question: Will writing a journal about the potentially stressful life experience of having an infant hospitalized for at least one week in the NICU reduce symptoms of psychological distress when compared to a waiting list control group?

Alternative Hypotheses: Writing a journal about the potentially stressful life experience of having an infant hospitalized for at least one week in the NICU will reduce symptoms of psychological distress when compared to a waiting list control group.

Null Hypotheses:Writing a journal about the potentially stressful life experience of having an infant hospitalized for at least one week in the NICU will not reduce symptoms of psychological distress when compared to a waiting list control group.

Statistical Sentence:

• Omnibus F (2,33) = 14.952, p = .000
• SCL-90-R F = 26.164, p = .000, Cohen's d = .805
• IES-R F = 17.817, p = .000, Cohen's d = .766.

Interpretation of the p value:

All 2 p-values in the main analysis were = .000, leaving no possiblity that the Null Hypothesis was true.

Sample Response: Sample Paper #4

Research Question(s)

1. Through systematic strategy instruction, will children with specific learning disabilities' test scores increase between pre and post test?
2. Through systematic strategy instruction, are children with specific learning disabilities able to achieve the same passing rate as general education students, on the writing portion of state competency exams?

Alternative Hypotheses:

1. Through systematic strategy instruction, children with specific learning disabilities' test scores will increase between pre and post test. (correct! the two group means will NOT be the same)
2. Through systematic strategy instruction, children with specific learning disabilities will achieve the same passing rate as general education students, on the writing portion of state competency exams. (Note from Dr. Barry - This student's response, as with many others, is reversed. What they list as the alternative should really be the null hypothesis and vice versa. I know it seems backward, but if you think about what we are trying to show in this study, we are looking for the two group means to be the same, which in statistical terms, is the null hypothesis, not the alternative.  This is tricky and confusing to students because in this instance, to show that kids with SLD can achieve what general education kids achieve, we are actually looking for their group means to be equal, which is what the null hypothesis represents. We are hoping that through instruction, we will make the two group means more similar by increasing the student scores in the disability class to be more similar to the state average. So we are actually hoping for a non-significant statistical outcome which would indicate that the two groups are more similar, i.e. kids with SLD are achieving similar pass rates as general education students. Think about how this student answered and why they get confused later on.)

Null Hypotheses:

1. Through systematic strategy instruction, children with specific learning disabilities' test scores will not increase between pre and post test. (correct! the two group means will be the same)
2. Through systematic strategy instruction, children with specific learning disabilities will not achieve the same passing rate as general education students, on the writing portion of state competency exams. ( again, this should be reversed, because of the nature of the null hypothesis. Statistically, it has to represent the two group means being equivalent.)

Statistical Sentences:

1. A t-test comparison between pre and post test scores for the treatment group was statistically signficiant, t (19) = -4.034, p = .001, Cohen's d of 0.920 and an effect size r of 0.418
2. A t-test comparison between group means of the treatment group and the average for the general education population of the same grade on the writing exam was not statistically insignificant, t (225) =1.269, p= .206, with a Cohen's d of 0.325 and an effect size r of 0.160.

Interpretation of the p value:

1. The p value of .001 which gives a 0.1% chance that the null hypothesis is true. We would reject the Null Hypothesis and Accept the Alternative.
2. The p value of .206 which gives a 20.6% chance that the null hypothesis is true. We would have to accept the Null Hypothesis?

Student Question to Dr. Barry:  p=.206 or 20.6% - which, as other students have indicated, is high. Shouldn't the number "p" be less than or between: .01, .05, and .1? I feel I must be misunderstanding because the paper is in support of the self-directed organizational strategy, and in general the anecdotal findings favor the approach, yet the statistical finding of p=20.6% speaks against it. Clearly I am failing to understand something critical here. Can someone help me understand?

Note from Dr. Barry back to student: To understand this, we need to think about the null and alternative hypothesis. In this case we are hoping for a null result, that there is no difference between our group means. If the null hypothesis is true, then our two groups are equivalent which means that the SLD students' mean score was indistinguishable from the general education student's mean score, which was the goal of the study. I know this one is tricky, but it makes you think about interpreting what the p value means relative to the research question.

Help from Lesson: This one is tricky because there are 2 tests going on. One that is a pretest versus posttest of 1 group. They do test the pre to post group means using a Paired Sample t-test and find a significant difference between the pre and post group means, rejecting the Null Hypothesis that the two group means were equal.

You should know by now that this pre/post comparison by it self is not enough to have confidence in the intervention. Why? Because there is no comparison group. Therefore, the authors went further and compared their posttest data with the state average. However, instead of starting out with two equal groups and seeing if our intervention had an effect by making the group means significantly different from each other, in this study, the two groups start out different and the researcher's goal was to make them the same. Remember that the Null hypothesis ALWAYS tests the probablity of the two group means being equal. So, in this case we would want to accept the null hypothesis if the intervention had an effect, which is exactly what we do.