Give three statistics used to analyze correlational research. For each, explain when it should be used and what kind of conclusion could be drawn.

Question

What will be an ideal response?

Answer 1

Ans: There are a number of questions used in behavioral sciences research and in correlational research specifically. If one is considering the sample as a whole, there are two approaches: correlation (such as Pearson’s r) and regression. For Pearson’s r, you are asking whether two measures given to the participants are related to each other. In particular, the data need to be interval or ratio level. The numbers range from ?1.0 to + 1.0. The bigger the number, regardless of whether positive or negative, means a stronger relationship between the two variables. If the relationship is positive, the variables vary in the same direction. If the relationship is negative, the variables vary in opposite direction (e.g., as one goes up, the other goes down). In a regression analysis, we want to see whether scores on one measure predict the scores on the second measure. Both measures would be interval/ratio data. The analysis produces a regression coefficient, b. There is also the standardized regression weight, beta, and the unstandardized regression coefficient, b. The larger the number, the more of variance in the second variable is explained by the first. A third example of a statistic used in correlational research is the chi square. In general, researchers think of chi square as answering questions about already existing groups. Both the independent variables are nominal data. The data are expressed as a chi square and the larger the number, the more the groups (expressed as the first variable) are different on the second variable.