SAGE Journal Articles

Click on the following links. Please note these will open in a new window.

Straus, Murray A. The National Context Effect: An Empirical Test of the Validity of Cross-National Research Using Unrepresentative Samples. Cross-Cultural Research 43(3): 183-205.

The objective of this study was to investigate whether results from cross-national studies using convenience samples which are not representative of the nation can provide valid cross-national comparisons. Analysis of data from the International Dating Violence Study (IDVS) of university students in 32 nations (n = 17,404) enabled 18 tests of concurrent validity and found an average correlation of .51 between variables measured by the IDVS measured and by nationally representative samples. Construct validity was also supported by 41 empirical tests. The concept of national context effects explains how samples that are not nationally representative can provide valid nation-to-nation differences. It was concluded that convenience samples that are not representative of the nation but are comparable across nations can provide valid tests of theories about differences between nations. Consequently, if a study can only be done using convenience samples, that should not deter proceeding.

Barratt, Monica J., Jason A. Ferris, and Simon Lenton. Hidden Populations, Online Purposive Sampling, and External Validity: Taking Off the Blindfold. Field Methods 27(1): 3-21.

Online purposive samples have unknown biases and may not strictly be used to make inferences about wider populations, yet such inferences continue to occur. We compared the demographic and drug use characteristics of Australian ecstasy users from a probability (National Drug Strategy Household Survey, n = 726) and purposive sample (online survey conducted as part of a mixed-methods study of online drug discussion, n = 753) using nonparametric (bootstrap) and meta-analysis techniques. We found significant differences in demographics and drug use prevalence. Ideally, online purposive samples of hidden populations should be interpreted in conjunction with probability samples and ethnographic fieldwork.