This study is concerned with the demographics and determinants of a test of crystallized (Gc) intelligence [1
] assessed in middle age. In the present longitudinal study, we used vocabulary task performance as a proxy for Gc. It has the advantage over previous studies on the determinants of adult IQ by not only having a large, representative adult sample of over 4000 adults, but also by being able to examine different intelligence measures assessed 32 years apart, from childhood to adulthood. We were able to do this together with a set of socio-demographic and psychological factors such as parental social status, locus of control, psychological distress, education, and occupation in influencing adult vocabulary task performance. This allowed us to explore a number of different hypotheses.
1.1. The Stability of Intelligence
There is considerable evidence of the stability of intelligence over time. There have been a number of different data sets examined, all of which have shown considerably high correlations between intelligence measured in childhood, mid, and late adulthood [2
]. One of the explanations is that general cognitive ability or intelligence is largely genetically determined [4
]. Newly emerging quantitative genetic designs, such as twin studies, have consistently shown that genetic influence on individual differences in intelligence is substantial [5
]. A recent study [6
] of over 4000 Vietnam veterans over an 18-year period using two tests (one verbal and one arithmetic) showed a stability coefficient of 0.79 for arithmetic and 0.82 for verbal ability, with an increase in the verbal ability (107.16 to 116.52).
It may therefore be expected that any and all valid measures of childhood intelligence have a significant and direct influence on adult intelligence, even with different and brief measures of intelligence.
1.2. Locus of Control and Intelligence
Locus of control is conceived of as a belief that a response will or will not influence the attainment of reinforcement [7
]. It is a ‘problem-solving’ generalized expectancy, addressing the issue of whether behaviors are perceived as instrumental to goal attainment, regardless of the specific nature of the goal or reinforcer. Locus of control is seen to influence the particular goal expectancy in any given specific situation depending upon the novelty and the ambiguity of the setting, as well as the degree of reinforcement that the individual has directly experienced in that setting [8
Various studies have looked at locus of control as a mediator variable between early childhood experiences and later effects in adolescence, such as anxiety and depression [9
]. Relatively fewer studies have looked at the relationship between locus of control and intelligence. In two early cross-sectional studies with small samples (n = 58 and n = 45) [11
] found intelligence and internal locus of control to be positively correlated among adolescents. A later study [13
] demonstrated that childhood intelligence was predictably related to locus of control, and that both factors were related to childhood behavioral intelligence.
In a recent study using a large sample [14
] it was found that childhood intelligence significantly predicted teenager internal locus of control (regression coefficient = 0.36, p
< 0.001). This study was able to examine whether teenager locus of control predicts adult vocabulary task performance.
1.3. Psychological Distress and Intelligence
There is evidence showing that low childhood intelligence is a risk factor of adult psychological distress [15
]. Psychological distress, in turn, is a significant predictor of functional memory [17
]. Further, lifetime depression has been found to have a significant effect on self-reported cognitive problems [18
], and anxiety has been found to have significant association with cognitive-task performance [19
This study used malaise as a measure of distress, which has shown to be relatively stable over time (Funrham and Cheng, 2015) [20
]. This study was able to examine whether psychological distress affects adult vocabulary task performance.
1.4. Parental Social Status, Education, Occupation, and Intelligence
Various studies have demonstrated the links between family socio-economic conditions, early cognitive ability, and later educational and occupational outcomes [21
]. Because intelligence is significantly heritable, and social class and intelligence are related, it is not surprising that parental social status is a significant predictor of children’s educational and occupational outcomes.
This study explored the effects of family social status, childhood cognitive ability, teenager locus of control, educational qualifications, and occupational prestige using a path model and drawing on data collected from a large representative population sample born in 1970 in the UK.
Based on the literature reviewed, six hypotheses were formulated: (H1) Childhood cognitive ability is significantly and positively associated with adult vocabulary task performance; (H2) Internal locus of control measured during teenage years is significantly and positively associated with adult vocabulary task performance; (H3) Educational and occupational achievement is significantly and positively associated with adult vocabulary task performance; (H4) Psychological distress is significantly and negatively associated with adult vocabulary task performance; (H5) Parental social status is significantly and positively associated with adult vocabulary task performance; and (H6) Childhood cognitive ability, locus of control, psychological distress, education, occupation, as well as parental social status are independently associated with the outcome variable.
The results of the current study show that all of the independent variables measured at different points in time, except gender and occupation, were significantly associated with adult vocabulary task performance. The SEM results in Figure 1
show that childhood cognitive ability, locus of control during teenage years, educational qualifications, and occupational prestige were significant and direct predictors of adult vocabulary task performance, whilst parental social status influenced the outcome variable mainly mediating through education.
There are various important points that can be drawn from this study. First, Figure 1
shows that almost half of the variance could be accounted for by four factors—namely childhood cognitive ability, locus of control, education, and occupation—three of which were measured 8–32 years earlier. Thus, ability and personality impact education, which affords a fine opportunity to increase crystallized intelligence and to guarantee a better and more cognitively challenging profession.
Second, the strongest predictor of the outcome variable was childhood cognitive ability measured at age 10 years (accounting for 31% of the total variance), indicating the stability of verbal ability—one of the main components of general cognitive ability or intelligence assessed 32 years apart. This confirms the literature established in the area [2
]. Due to the fact that we were only using relatively short test at age 42 years (4 min), the correlations over 30 years apart were considerably high.
Third, teenager locus of control was significantly influenced by childhood cognitive ability and parental social status, which in turn had direct positive effects on education and occupation, as well as on adult vocabulary task performance. This suggests that more intelligent children have a more instrumentalist, internal locus of control as teenagers, which in turn influences educational and occupational success and verbal intelligence in adulthood. Locus of control was also a significant predictor of psychological distress, suggesting that mental health, in part, might be maintained by internal locus of control beliefs and a sense of control in one’s life.
Fourth, the present study confirms the significant and direct effects of parental social status and childhood cognitive ability on education and occupation, which are in line with the literature in the area [22
Although psychological distress was significantly correlated with adult task performance, it was not an independent predictor of the outcome variable. It might be related to the levels of cognitive tasks. The literature on the relationship between personality and intelligence has always demonstrated a negative correlation between neuroticism and IQ—because the test anxiety associated with the trait impairs test performance on timed tests—as found in a study that indicated both neuroticism and psychological distress are significant predictors of adult functional memory [17
Fifth, the present study indicates that in order to maintain a good verbal ability in adulthood, it would be helpful to enhance children’s internal locus of control through early education in both family and school settings. The study also emphasizes the importance of higher education and lifelong learning. Although it might be difficult to change childhood cognitive ability, since childhood cognitive ability is significantly associated with parental social status, it is helpful to improve economic conditions in disadvantaged families and regions. This has long been the mantra of educationalists [21
Like all studies, this study had limitations. As with all research using cohort studies, the variables used in the study were constrained by the availability of the data. Another limitation is the attrition of respondents over time. Since sample attrition is greatest amongst individuals in more deprived circumstances, our results may, thus, be a conservative estimate of the long-term influence of social inequalities experienced during childhood. It would be desirable to replicate these findings on a more robust battery of intelligence tests in adulthood, including testing possible different pathways to adult fluid and crystallized intelligence [1