SOC 341-01/51 (Kuechler)
Fall 1997
Select one of the three topics below:
Find two scholarly articles dealing with the selected topic that include a precise description on how the key variable ("spanking", "marital happiness", or "alcohol abuse") was measured.. For these two articles
Then
Well, it appears that a lot of students had trouble with this one. I guess, few took my hint seriously:
Reread chapters 1 and 2 with respect to measurement validity. Keep your discussion to this particular issue, do not summarize the whole study, the findings, etc.
Quite a few students addressed other validity issues, in particular "causal validity" (e.g., whether a link between two variables had been established correctly) and "external validity" (e.g., whether or not the sample was representative of all Americans). Important issues, but not what the assignment called for. For this one, you had to pick one of the concepts above and check on the concepts was measured in two different studies. The rest of study (article) was of no interest for this assignment. So, you skip over most of the piece, the sampling, the findings, etc. But you needed to locate the details about the measurement of the concepts. I will walk you through (mostly using one particular student's submission) the details using marital happiness as an example. Before, however, two general remarks:
Sample submission
1. Bibliographic details:
Glenn, N.D., and S. McLanahan (1982). "Children and Marital Happiness: A Further Specification of the Relationship". Journal of Marriage and the Family 4:63-72.
White, L.K., A. Booth, and J.N. Edwards (1986). "Children and Marital Happiness: Why the Negative Correlation?" Journal of Family Issues 7:137-147.
Comment: Make sure to include all important details. In several submissions either journal name or name(s) of the author(s) were missing.
2. Description of the measure(s) used:
Glenn study:
Marital happiness was gauged by the following question: "Taking things altogether, how do you describe your marriage? Would you say that your marriage is very happy, pretty happy, or not too happy?" The responses were scored from 0 ("not too happy") to 2 ("very happy") and treated as an interval scale.
White study:
Marital happiness was measured using an 11-item Likert-type scale (agree-disagree scale -- usually ranging from 1 to 5) with seven items about specific aspects of marriage (extent of understanding, agreement with spouse, sexual relationship, spouse's faithfulness, amount of love and affection received, spouse as someone who takes care of things around the house, spouse as someone to do things with) and four global satisfaction items (overall happiness of marriage, rating of own marriage compared to others, strength of love for spouse, and whether marriage is getting better or worse). Answers indicating happiness were scored 5, so that the higher the score the happier the marriage.
3. Assessment of measurement validity
Glenn study:
Purely subjective approach. Definition of happiness is left to individual who answers it. If one assumes that people differ in what makes them happy, this is a reasonable approach. However, there is pressure (social desirability) to portrait oneself in a positive way, so answers may not be truthful. They may also vary given the specific circumstance of when and where the question was asked. For example, if someone has had a major fight with his/her spouse, he/she is not likely to say "very happy". However, a week later, after the couple has kissed and made up, he/she might say "very happy". A serious reliability problem, and no validity without reliability.
White study:
The much larger set of questions concerning marital happiness make people really think about their marriage. The answers are less likely to be influenced by specific circumstance (like a current or recent fight). So, the measurement appears to be as fairly reliable. The 11 items all have face validity, that is they deal with aspects that most people would agree are relevant in determining marital happiness. Summing up the individual scores may be seem a bit problematic. For example, a "1" on sex and a "5" (taking care of things around the house) amounts to the same sum score of 6 as if the answers were the other way around (great sex, no housework). But, again, different strokes for different folks.
4. Compare and contrast
The measurement in the White study has much higher validity than the Glenn study. Something very close to the single item used by Glenn is included in the 11-item set used in the White study which has no apparent reliability problems and good face validity. So, I certainly would pick the approach used by White and colleagues.
5. Copies (excerpts) of articles
Most students got this right and only submitted the one or two pages relevant for this assignment. However, quite a few failed to clearly mark what section in particular provided the base for their discussion. But, then, there was quite a bit of confusion about what to do in the first place.