Modern education may be a possible explanation for the Flynn Effect (FE). However, the precise mechanisms by which modern education influences the rise in IQ test scores is not altogether clear. De Boeck (personal communication, January 22, 2017) suggests, that the FE should be thought of in terms of automated processing, wherein faster and more accurate responses, rather than controlled ones, facilitate higher test scores. Due to the fact that many IQ test items already bear a close resemblance to school subjects, we are able to conclude that modern education enhances the capacities of students to take tests that require such kind of fast intellectual reactions. The aim of the current paper is to investigate the differences in test-taking speed among several FE samples of 13-14 year old students in Estonia. There was a 2-year difference in education levels between the cohorts. We found that the younger cohort finished their IQ tests more quickly than the older one, and experienced less time pressure as well.
The two cohorts of Estonian students (1933/36, n = 890; 2006, n = 913) were compared using an Estonian adaptation of the National Intelligence Tests (NIT), which is a test battery consisting of 10 subtests. The significant differences in test-taking speeds between the cohorts were recorded. Based on the data we were able to discern a positive FE. Test-taking speed was calculated according to guidelines developed by Stafford (1971), who proposed that test-taking speed is determined by the proportion of unattempted items in the total number of errors (speediness coefficient, SQ). Using the Minimal Restriction Speediness Model (MRSM; Estrada et al.,2017) we were able to separate the level (the variability of the test scores attributable to the differences in ability that the test is designed to measure) and the speed factors of the test scores. Analysis of all of the subtests showed that the NIT was more of a speeded test for the older cohort than it was for the younger cohort (SQ-s in range .24 - .49 and .10 - .25 correspondingly). The sharpest difference between the cohorts occurred in the Comparsion subtest (SQ-s .44 vs .17 correspondingly). The older cohort experienced more time pressure and frequently was unable to complete the subtest. The SQ-s of the subtests’ items and item’s loadings on the speed factor were strongly correlated: the strongest correlation, r =.94 was in the Same - Different subtest. The strong correlation between the two different speed estimates demonstrates that they utilize the same speed processing construct.
Based on the speed perspective we are able to conclude that the FE is not solely indicative of differences in mental levels, or the ability to solve mental items; it can also demonstrate the cohort’s readiness to solve the presented items quickly. As education has made IQ-test-like items (computation, arithmetical problems, vocabulary, synonyms-opposites, sentence completion, information etc.) more common and familiar to students, it is natural that high speed answering would become more common as well. In this way we can begin to explain some of the connections and the influences of education on the FE.
