The psychometric equivalency of Internet vs. in -person research
The current study addressed the question of whether test data can reliably be gathered on the Internet. The Internet gives researchers a large population from which to draw more representative subjects than do the typical samples of convenience currently used in research. To this end, a measure that had previously been created and validated via a standard, in-person format, the Sexual Boredom Scale, was presented via the World Wide Web, along with five additional scales used to validate the original measure. Subjects were sorted demographically. A subset of 533 participants that matched the original sample on age, gender, and race was identified. A correlation matrix of 13 scores based on the 6 tests was calculated for the scores from the matched sample of 533 subjects. This matrix was compared via structural equation modeling to the correlational matrix from the original study and goodness of fit scales were generated. The goodness of fit results indicated that the Internet sample replicated the original, paper-and-pencil sample. In addition, coefficient Alpha values were calculated for the matched sample, yielding almost identical reliability coefficients to the original test values. Factors such as the computer as intermediary and uncontrollable administration settings did not appear to have any effect on results. Demographic data suggested that while the population of the Internet is not yet equivalent to the U.S. Census, it is moving in that direction and is most certainly more representative of the census than is the typical college sample. In addition to the matched sample, from which the actual matrix comparison was made, the web site gathered a total sample of almost 2400 subjects in 8 months without remuneration. These results suggest that data collection on the Web is (1) efficient, (2) cost effective, (3) representative, and (4) reliable.
Meyerson, Paul Eric, "The psychometric equivalency of Internet vs. in -person research" (2001). ETD Collection for Fordham University. AAI3017558.