Hypothesis testing - case studies using research reports
Studying research reports to extract the basic information about the study design, analysis and conclusions can help broaden one's perspective on how hypothesis testing is used to do research in the real world. Research reports can be dense descriptions of esoteric concepts using complex study designs to investigate a number of issues. Don't worry too much about understanding every bit of the report (in particular, the topic may be something completely unfamiliar and thus hard to understand or the analyses may include tests that you've never hear of). The goal is to carve out an understanding of the 4 steps involved in hypothesis testing as implemented in the study:
- Null and alternative hypotheses
- Collecting and summarizing data
- Assessing the evidence provided by the data
- Making conclusions in the context of the original hypotheses
Choose a publicly available research report. Choose a report from the listing below (organized according to statistical test), or identify another suitable report. Read and review the report. Summarize your findings for each of the four steps, Use the questions listed in the next section as a guide.
Guiding questions to summarize the 4 steps in hypothesis testing
- What is the topic of this research?
- Why is this research of interest?
Step 1: State the appropriate null and alternative hypotheses, Ho and Ha.
- What are the hypotheses being tested in the study?
Step 2: Obtain a random sample, collect relevant data, and check whether the data meet the conditions under which the test can be used. (random sample, normal distribution applies--check summary stats) If the conditions are met, summarize the data by a test statistic.
- What is the sample(s) used in the study?
- What test will be used?
- Does each sample meet the conditions for using this test?
- What is the value(s) of the test statistic?
Step 3: Find the p-value of the test
- What is the p-value of the test statistic.
Step 4: Based on the p-value, decide whether or not the results are significant and draw your conclusions in context.
- What decision is made with respect to Ho and Ha?
- Given that decision, what conclusions were made based on the research?
- How is this research limited in what it can conclude?
Links to research reports
The reports provided are organized by statistical test, for use by students as they learn about the different tests.
Independent samples and matched pairs t-test
In some cases the t-test will be used to evaluate only one aspect of the study. Focus on just the aspect for which the t-test is used.
- A Rose by Any Other Name . . . : Color-Naming Influences on Decision Making by Jeanine L. Skorinko, Suzanne Kemmer, Michelle R. Hebl, David M. Lane
- A cross-sectional analysis of video games and attention deficit hyperactivity disorder symptoms in adolescents by Philip A Chan and Terry Rabinowitz
- Striking a Responsive Chord: How Political Ads Movitate and Persuade Voters by Appealing to Emotions by Ted Brader [This article is somewhat long; you may want to focus on only a portion of this study.]
In some cases the one-way ANOVA will be used to evaluate only one aspect of the study. Focus on just the aspect for which the ANOVA is used.
- Effectiveness of Negative Political Advertising by Won Ho Chang, Jae-Jin Park, and Sung Wook Shim
- Explanatory Style of Secondary Vocational Educators by
Helen C. Hall and Bettye P. Smith
Finding a suitable research report
Use a Google scholar search including the name of the test you are interested in (put this phrase in quotes if more than one word, e.g., "t test") and one or two topic words. Links to full-text versions of studies, when available, are listed in the right hand column of the search results page.