l23 - Scale Construction(Very Brief Overview This handout...

This preview shows pages 1–3. Sign up to view the full content.

Scale Construction (Very Brief Overview) —Page 1 Scale Construction (Very Brief Overview) This handout draws heavily from Marija Norusis’s SPSS 14.0 Statistical Procedures Companion. See her chapters 18 (Reliability Analysis) and 17 (Factor Analysis), as well as Hamilton’s (2004) ch. 12 (Principal Components, Factor, and Cluster Analysis) for much more information. I’ve adapted Norusis’s example from ch. 18 to show how the analyses could be done in both Stata & SPSS. As we have seen, individual items often suffer from random measurement error; and when several items all tap the same underlying concept, including them all separately in a regression can lead to problems of multicollinearity and unnecessarily complicated results. Therefore, it is sometimes desirable to create scales out of items. When individual items all tap the same concept, a well constructed scale will be more reliable than each item individually and will also be more parsimonious. When constructing scales, naturally we want to know how good the scale is. How reliable is the scale? What items belong in the scale? Both theoretical concerns and empirical results should guide you in scale construction. The General Social Survey asked 1,334 respondents questions concerning anomie, i.e. the breakdown of social norms. It consists of 9 statements to which people answer either agree (coded 1) or disagree (coded 0). In SPSS, the scale = the number of statements with which they agree; in Stata that score is divided by the number of items in the scale. Here are the descriptive statistics: . use http://www.nd.edu/~rwilliam/stats2/statafiles/anomia.dta, clear . describe an* storage display value variable name type format label variable label ------------------------------------------------------------------------------- anomia1 byte %8.0g anomia1 next to health, money is most important anomia2 byte %8.0g anomia2 wonder if anything is worthwhile anomia3 byte %8.0g anomia3 no right and wrong ways to make money anomia4 byte %8.0g anomia4 live only for today anomia5 byte %8.0g anomia5 lot of average man getting worse anomia6 byte %8.0g anomia6 not fair to bring child into world anomia7 byte %8.0g anomia7 officials not interested in average man anomia8 byte %8.0g anomia8 don't know whom to trust anomia9 byte %8.0g anomia9 most don't care what happens to others . sum an* Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- anomia1 | 3975 .3116981 .4632455 0 1 anomia2 | 3975 .4125786 .4923601 0 1 anomia3 | 3975 .247044 .4313468 0 1 anomia4 | 3975 .4445283 .4969759 0 1 anomia5 | 3975 .5836478 .4930154 0 1 -------------+-------------------------------------------------------- anomia6 | 3975 .3889308 .4875689 0 1 anomia7 | 3975 .6332075 .4819898 0 1 anomia8 | 3975 .7177358 .4501578 0 1 anomia9 | 3975 .5554717 .4969759 0 1

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Scale Construction (Very Brief Overview) —Page 2 As you can see, only 25% agree with the statement “there are no right or wrong ways to make
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 02/29/2012 for the course SOC 63993 taught by Professor Richardwilliams during the Spring '11 term at Notre Dame.

Page1 / 6

l23 - Scale Construction(Very Brief Overview This handout...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online