Reliability: Correctly use terms to define the types of evidence for reliability reported in the review articles, and provide the specific numerical values of the reliability statistics. If no reliability data are provided, then explain what type of evidence for reliability you would need in order to fully evaluate the test.

Validity: Correctly use terms to define the types of evidence for validity reported in the review articles, and provide the specific numerical values of the validity coefficients. If no validity data are provided, then explain what type of evidence for validity you would need in order to fully evaluate the test.

The reading to evaluate is below

Review of the Miller Marriage Satisfaction Rating Scale by FRANK M. BERNT, Associate Professor of Health Services, St. Joseph's University, Philadelphia, PA: DESCRIPTION. The Miller Marriage Satisfaction Rating Scale is a 27-item self-report scale designed to measure marital satisfaction. Each item includes a statement about one facet of marriage or family, to which one responds using a 4-point rating scale (a lot, moderately, somewhat, not at all). The author proposes that the scale can be used by individuals who are "dating, living together, or who are married." Its apparent appeal is that it is self-administered and provides a nearly instantaneous score to anyone interested; the only requirement is internet access and a valid credit card. The scale yields a single percentile score, indicating one's position relative to other couples who have taken the scale. Who those other couples are is not made clear in any of the materials made available for review. DEVELOPMENT. No technical manual is available for the scale; instead, a very brief description of the scale is provided on the author's web page. There are no descriptions of item generation and selection, of pilot testing, or of standardization or norm-referencing procedures. TECHNICAL. Information about the reliability and validity of the scale is presented on the author's web page. Test-retest reliability is reported to be .92; however, the author does not cite studies done, describe samples used, or define the intertest interval. The discussion on the web page seems to suggest that the split-half reliability for the test is less than .70. Given the relatively large number of items on the scale, it would seem that more careful item selection based upon an item analysis would yield a stronger internal consistency estimate. If its original appeal was its availability on the internet, this is overshadowed by the existence of many alternative measures of marital satisfaction now available on the internet that are much less expensive (often free) and in most cases also well-documented with respect to validity and reliability information. The author's discussion of validity skirts the issue of whether this test has demonstrated validity or not. The author admits that "a large number of additional studies will be needed" to determine the utility of his scale; what is left unsaid is that not a single study is cited in support of the scale's validity. At the same time, he is very optimistic that "future research will prove that the Miller Marriage Satisfaction Scale is useful for therapy, research and personal information" (retrieved September 26, 2001 from www.metadevelopment.com/faq/marriagesatisfaction.html). COMMENTARY. It is the reviewer's strong opinion that this test, at its present level of development, is not suitable for public use. Reliability and validity information are scant or nonexistent, falling far short of published Standards for Educational and Psychological Testing (AERA, APA, & NCME, 1999). The user pays $35 for a computer-administered survey (single-use) and receives in return a percentile score indicating relative position in a norming sample that is completely anonymous and amorphous; the size and demographic characteristics of the sample, as well as the sampling method used, are not available. SUMMARY. The Miller Marriage Satisfaction Rating Scale, unfortunately, has very little to recommend it. Until the author provides the most basic and fundamental information outlined in the Standards for Educational and Psychological Testing (AERA, APA, & NCME, 1999), there are many other scales-for example, the Marital Satisfaction Inventory-Revised, (Snyder, 1998), the Dyadic Adjustment Scale (Spanier, 1976), the ENRICH Marital Satisfaction Scale, and the Locke-Wallace Marital Adjustment Scale (Locke & Wallace, 1959)-which are widely used, well-documented, and hence highly recommended over the scale in question. REVIEWER'S REFERENCE American Educational Research Association, American Psychological Association, & National Council on Measurement in Education. (1999). Standards for educational and psychological testing. Washington, DC: American Educational Research Association. Locke, H. J., & Wallace, K. M. (1959). Short marital adjustment and prediction tests: Their reliability and validity. Journal of Marriage and the Family, 21, 251-255. Spanier, G. B. (1976). Measuring dyadic adjustment: New scales for assessing the quality of marriage and similar dyads. Journal of Marriage and the Family, 38, 15-28. Snyder, D. (1998). Marital Satisfaction Inventory-Revised. Los Angeles: Western Psychological Services.

Review of the Miller Marriage Satisfaction Rating Scale by CINDY I. CARLSON, Professor of Educational Psychology, University of Texas at Austin, Austin, TX: DESCRIPTION. The Miller Marriage Satisfaction Rating Scale (MMSRS) is a 27-item test administered by computer. Items are to be answered to indicate how much the statement is characteristic of the respondent. All items have the same four response options: a lot, moderately, somewhat, and not at all. The items form a scale that provides a single score. The score is given as a percentile rank. Respondents may place themselves within four levels of scores that range from minimal marriage satisfaction to high marriage satisfaction. The measure is designed to be completed and scored on the internet: http://www.metadevelopment.com. Results are immediately available upon completion of the scale on a single-page printout. A Visa or MasterCard is required to use the test. The cost of completing the test is $35.00. DEVELOPMENT. The MMSRS was published in May 1997. The author provides no information about test development on the website. The scale is purportedly developed from theory and research on marital satisfaction; however, the theory is not specified and no references are provided. Neither is information provided regarding pilot testing, test revision, internal consistency of the scale, or the characteristics of sample(s) on which the measure was tested. The measure lacks a manual and norms. There is no guidance for the user regarding populations for which the measure is appropriate beyond couples that are dating, living together, or married. It is not clear whether the measure has been used with or would be appropriate for homosexual as well as heterosexual couples; nor is there any information about the applicability of the scale to ethnically and socially diverse populations. Items are concise and clearly written. A review of the items, however, suggests that items tap at least two different perspectives of marriage. Whereas 12 items reflect characteristics of the respondent as a partner in the marriage and begin with the pronoun "I," the remainder refer to characteristics of the marriage. The response options are more appropriate for the items that reflect characteristics of the respondent as a partner in the marriage, and it is difficult to use the response options with the items that reflect marital quality. For example, the item "Neither partner is romantically involved with another person outside of the marriage" is not easily answered with a lot or moderately. TECHNICAL. There is no information on standardization of the measure. It is impossible to determine the appropriateness of this scale for different gender, ethnic/culture, or sexual orientation groups. The author reports a test-retest reliability coefficient of .92, which is very good; however, there is no information regarding the length of time between the test-retest or the sample size. The author provides several split-half reliability coefficients, but information is inadequate to determine the usefulness of these values. No data on internal consistency are provided by the author. This is viewed as a major shortcoming of the measure given the obvious variation in item perspective between respondent as a marital partner and the perceived quality of the marriage. Moreover, the lack of internal consistency data is inconsistent with the express purpose of the measure to "permit the individual to look at different areas of their marriage." There is no indication that this measure taps different areas of marriage. Rather it purports to measure a single construct and provides no statistical support for that assertion. There is no reported research on the validity of scores from the MMSRS. COMMENTARY. Strengths of the MMSRS include ease of use, readability, and relatively inexpensive cost. Weaknesses include the lack of empirical support for the psychometric quality and treatment utility of this measure. There is no evidence of adequate test development, appropriate norms or standardization, or validity of the measure for its purported purpose. SUMMARY. The MMSRS may be a very convenient measure for individuals or couples to use to quantify their marital satisfaction. The lack of information on test development and psychometric quality suggests that both therapists and researchers would be better served by utilizing marital satisfaction measures that have been more carefully developed and supported by a program of research.