1



Examples of Published Research With Analysis Including a Bivariate Correlation/Regression AnalysisThese summaries were provided by previous studentsCeylan, R. & Aral, N. (2007). An examination of the correlation between depression and hopelessness levels in mothers of disabled children. Social Behavior and Personality, 35, 903-908.Subjects: The subjects were mothers with disabled children between the ages of 5 and 7 at private schools also known as the First Education Schools of the Ministry of Education.Variables studied: Measure depression and hopelessness levels in the subject mothers with disabled children using the Beck Depression Scale as well as the Beck Hopelessness Scale.Design: It was an observational design as the researchers used surveys to measure depression and hopelessness levels.Conclusion: There was a significant positive relationship between levels of depression and hopelessness.Quote: The results showed that there was a positive and significant correlation between depression and hopelessness (r = 0.72, p < .001). This suggests that when the depression levels of mothers increase, so do their hopelessness levels, and that greater hopelessness levels tend to result in higher levels of depression.Critique: The design and results appear pretty straightforward. The researchers didn’t include N in their summary statement. They also didn’t include a 95% confidence interval for r.Graduate student Tim Bibo found an example of research comparing slopes: Cameron, H. & Robertson, A. (1997). The colon in medicine: Nothing to do with the intestinal tract. British Medical Journal, 315, 1657-1658.The authors' summary statement was: “Comparisons between journals showed a significant difference in the rates of increase between the New England Journal of Medicine (slope 0.006 (95% confidence interval - 0.036 to 0.048)) and both the Lancet (slope 0.35 (0.17 to 0.53); t= 3.74, df = 1, P < 0.001) and the BMJ (slope 0.72 (0.35 to 1.09); t = 3.72, df = 1, P < 0.001). The slope estimates for the BMJ and the Lancet were not significantly different.”What Tim has reported here is an example of a test of the null hypothesis that the slopes for predicting Y from X are the same in one population as in another. When we study multiple regression in PSYC 7431 we shall study this topic in a more sophisticated fashion, the Potthoff analysis.There has been an increase in the rate of use of colons in journal titles, especially after an article years ago indicated that articles with colons in their titles are more likely to be favourably reviewed than are articles without such colons. The authors of the article Tim reviewed compared the rate of growth (slope = change in number of colons divided by passage of time) in use of colonic titles across three journals (three pairwise contrasts). ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download