1 Chapter 5 Descriptive Research © 2005 Thomson/South-Western
2 Descriptive Studies Cross- Sectional Omnibus Panel Longitudinal Sample Survey True Panel Figure 1: Classification of Descriptive Studies
3 Longitudinal Versus Cross-Sectional Data Longitudinal Data Cross Sectional Data Allows the investigation of a great many relationships Allows collection of a great deal more classification information from respondents Allows longer and more exacting interviews Produces fewer errors in reporting past behavior because of natural forgetting Produces fewer interviewer-- interviewee interaction errors Produces fewer errors due to respondent’s behavior being affected by the measurement task Tends to produce more representative samples of the population of interest Allows turnover analysis if panel is a true panel
4 Number of Families in Panel Purchasing Each Brand Brand Purchased During First Time Period, t 1 During Second Time Period, t 2 A B C D150 Total1000
5 Number of Families in Panel Purchasing Each Brand in Each Period During Second Time Period t 2 During First Time Period, t 1 Bought A Bought BBought C Bought D Total Bought A Bought B Bought C Bought D Total
6 Number of Families in Panel Purchasing Each Brand in Each Period During Second Time Period t 2 During First Time Period, t 1 Bought ABought BBought CBought D Total Bought A Bought B Bought C Bought D
7 Set-Up of Cohorts Age C5C6C7C C4C5C6C C3C4C5C C2C3C4C5 50+C1C2C3C4 C1-cohort born before 1920 C2-born C3-born C4-born C5-born C6-born C7-born C8-born
8 %Purchasing Life Insurance Age
9 Research Realities 2: Comparison of Responses of the Market Facts Mail Panel and a Randomly Selected Telephone Sample Source: “Mail Panels vs. General Samples,” Research on Research 59 (MarketFacts.com).
10 Figure 3: Television Viewing Habits of Children