e-Articles (Chapter 6)
Here are some full-length research articles that illustrate the presentation of estimated standard errors, point estimates, and confidence intervals. To view any article, simply click on its title. (NOTE: No claim is made that these articles are perfect in all respects. By carefully reviewing them, you will hone your skills at being able both to decipher and to critique statistically-based research reports.)
Presentation of values for the SEM (i.e., the estimated standard error of the mean) help readers remember that the sample means, computed for each of 2 groups on each of several dependent variables, would likely be different if based on a different sample. See Tables 2 and 3.
Illustrates how 95% confidence intervals can help with the interpretation of sample statistics. In this study, CIs were built around the participants' mean age, height, and weight, their rates of injuries, and the ratio of such rates based on 2 different samples. See the 1st paragraph of the "Data analysis" portion of the "Methods" section, as well as Tables 1, 2, 3, 5, and 6.
In this research report, a comparison was made among 4 ethnic groups in terms of their use of public parks. The percentage of each group using parks for biking was shown as a bar in a bargraph. The same bargraph showed each group's usage for each of several other things (e.g., swimming, for picnicking). The researchers put an "error bar" into their graphic summary to indicate the 95% confidence interval for each ethnic sample's percent-usage of each possible park activity.
Illustrates how confidence intervals can be places around point estimates of reliability. See the 2nd and 3rd sentences of the "Results" portion of the "Abstract," the 2nd and 3rd paragraphs of the article's "Results" section, and the 1st paragraph of the article's "Discussion" section.
Copyright © 2012
Schuyler W. Huck