StatCrunch logo (home)

Report Properties

from Flickr
Created: Feb 1, 2018
Share: yes
Views: 332
Results in this report
Data sets in this report
Need help?
To copy selected text, right click to Copy or choose the Copy option under your browser's Edit menu. Text copied in this manner can be pasted directly into most documents with formatting maintained.
To copy selected graphs, right click on the graph to Copy. When pasting into a document, make sure to paste the graph content rather than a link to the graph. For example, to paste in MS Word choose Edit > Paste Special, and select the Device Independent Bitmap option.
You can now also Mail results and reports. The email may contain a simple link to the StatCrunch site or the complete output with data and graphics attached. In addition to being a great way to deliver output to someone else, this is also a great way to save your own hard copy. To try it out, simply click on the Mail link.
Week 4 Report: Sleep Study
Mail   Print   Twitter   Facebook

Introduction: Insomnia is a condition which inhibits one's ability to sleep. According to DiBonaventura, Richard, Kumar, Forsythe, Flores, and Moline, insomnia sufferers have lower productivity at work and suffer from various heatlh detriments related to their chronic lack of sufficient sleep (DiBonaventura, Richard, Kumar, Forsythe, & Moline 2015). Many methods can be utilized to aid in both falling asleep and staying asleep. This study investigates a new nonprescription sleeping pill and it's impact on the average time it takes an individual to fall asleep compared to a placebo. 

Methods: The self-selected sample included 100 individuals who regularly suffer from insomnia. Participants took one pill one night and the second pill one night a week later. Some participants were given the new sleeping pill first, while others received the placebo pill first. A device was provided to each participant to determine the time it took to fall asleep. Interval/ratio data was being measured and the sample size of 100 allowed researchers to assume that normality conditions were met, so a matched-pairs t-test was utilized to determine if the new drug increases the amount of sleep. Measuring the same individual's response to both the pill and placebo allowed for the removal of person-to-person variability. A 95% confidence interval was obtained to determine if the average time to fall asleep with the new pill was less than the average time to fall asleep with the placebo. 

Analysis:The matched-pairs t-test yielded a low p-value (<0.0001), indicating that there is a difference in the time it takes to fall asleep between the drug and placebo (drug-placebo). The mean difference is negative, which shows that on average the pill causes one to take more time to fall asleep compared to the placebo.

Result 1: Matched Pairs t-test for week 4 report   [Info]
Paired T hypothesis test:
μD = μ1 - μ2 : Mean of the difference between Drug and Placebo
H0 : μD = 0
HA : μD ≠ 0
Hypothesis test results:
DifferenceMeanStd. Err.DFT-StatP-value
Drug - Placebo-1.560.2637415399-5.9148818<0.0001


The 95% confidence interval obtained includes 0 in its values, so there is no evidence that the average time it takes to fall asleep with the pill is less than the time it takes to fall asleep with the placebo. 

Result 2: Two sample T confidence interval for week 4 report   [Info]
Two sample T confidence interval:
μ1 : Mean of Drug
μ2 : Mean of Placebo
μ1 - μ2 : Difference between two means
(without pooled variances)

95% confidence interval results:
DifferenceSample Diff.Std. Err.DFL. LimitU. Limit
μ1 - μ2-1.561.0352055196.41794-3.60154450.48154453

Conclusion: Individuals struggling with insomnia may be eager to try new interventions and medications to help alleviate their condition, thus improving quality of life. The goal of this study was to determine if the new pill lessened the average amount of time to fall asleep. While there are a variety of factors affecting the time it takes to fall asleep, measuring each participant's average time to fall asleep on both the pill and placebo assisted in eliminating some of this variability, improving statistical power. The matched-pairs t-test indicated that those taking the pill do not fall asleep faster than when taking the placebo. The 95% confidence value obtained, containing both negative and positive values, rejects the hypothesis that the new pill would lessen the average time to fall asleep. Unfortunately, there is no evidence that the new sleeping pill in this study yields less time to fall asleep compared to the placebo group.


DiBonaventura, M., Richard, L., Kumar, M., Forstythe, A., Flores, N.M., & Moline, M. (2015). The association between insomnia and insomnia treatment side effects on health status, work productivity, and healthcare resource use. Plos ONE, 10(10).

HTML link:
<A href="">Week 4 Report: Sleep Study</A>

Want to comment? Subscribe
Already a member? Sign in.
By nku.dr.nolan
Feb 9, 2018

Hi Rosemary - Unfortunately I'm running short on time this week, so I want to preface my comments by saying that everything is getting a very cursory review. Please make sure that you have taken a look at the solutions for Week 4 statcrunch, and if you find discrepancies between your work and the solutions that you don't understand, please send an email to ask about them.

It looks like you mistakenly used the independent samples test for your CI (it was correct to use matched pairs. Check out the solution on this as the result is very interesting, especially in thinking about how to use it clinically (or not).
By julie.graessle
Feb 5, 2018

Very well designed report. I find it interesting that the matched pairs T Test revealed a mean difference between the two groups but the confidence interval lacked the ability to confirm a difference. This shows that although a mean difference exists between the two groups, the clinical impact and importance may not be as easy to determine.
By kyleigh.odom711
Feb 3, 2018

As someone who has trouble falling asleep, it is good to be able to learn how to interpret the results of a study involving new sleeping medications. I see that the hypothesis test did indicate that there was a difference in the time to fall asleep, but when you performed the confidence interval, there was not that big of a difference. I can see where some pharmaceutical companies may want to use just the result of a hypothesis test to support their interests.

Always Learning