Lynn and Hampson (1975) and recently Lester (2000) suggested that national level of extraversion (E) and neuroticism (N) could be assessed by using certain national indicators like suicide and divorce rates. In this study, Lynn and Hampson’s (1975) and Lester’s (2000) models were assessed by using 1990s data. Although Lynn and Hampson’s original N and E factors correlated with EPQ N (r=0.48) and E (r=0.60) scores as hypothesised, the factor structure did not recur in the 1990s data. Factor analysis supported Lester’s (2000) model of N and E. However, the factor loadings for EPQ N and E were rather modest (0.49 for N and 0.36 for E). It was concluded that more research with larger datasets and new social indicators are needed before social indicators could replace the EPQ in comparisons of national characteristics of nations.
Cross-cultural studies of national differences in personality are usually conducted by collecting personality questionnaire data in a set of countries. The objective of these kind of studies is to investigate if the theory (e.g. three factor structure of personality) emerges universally or if the instrument (e.g. Eysenck Personality Questionnaire) can be applied in different cultures and language areas. Unfortunately, cross-cultural studies of personality based on questionnaire data often have serious methodological problems. For example, translation of the instrument to several languages and establishment of equality of measurements might be difficult. Measurement errors and reliability of the measurements may vary between the countries. In addition to requirements related to the instrument, cross-comparisons require the samples collected in the countries to be identical in terms of demographic background and sampling strategy. Still, many other confounding factors cannot be controlled.
As early as in 1971, Lynn suggested that social indicators collected by national statistical agencies could be used for cross-national comparisons of personality. The social indicators have many attractive features compared to traditional personality questionnaire answers. Social indicators are not based on self-reports of behaviour or preferences, but are based on more objective measures (e.g. liver cirrhosis fatalities instead of self-reported alcohol use). It can be therefore assumed that social indicators are practically free of social desirability bias. In addition, social indicators are calculated from the whole population, i.e. people living in one clearly defined region, not from a sample. In this way, studies based on social indicators avoid problems caused by incomparable samples. Finally, social indicators are usually clearly defined and the measurement follows a standard procedure. Since social indicators are not measured with personality questionnaires, translation problems and issues related to cultural adaptation do not bias the measurements.
In his first study, Lynn (1971) compared the anxiety levels of industrialised nations by using such indicators as suicide, alcoholism and coronary heart disease rates. Lynn and Hampson, 1975 and Lynn and Hampson, 1977 extended the methodology to measurements of personality, namely, neuroticism and extraversion. Based on factor analyses of social indicators from 18 industrialised countries, they were able to identify two uncorrelated components, which they interpreted as neuroticism and extraversion (Lynn & Hampson, 1975). Later, Lynn and Martin (1995) used factor analysis to study the relationships between certain social indicators and EPQ N, E, and P scores. Low suicide and high homicide rates loaded on the same factor with extraversion whereas alcoholism was related to neuroticism. Other indicators from Lynn and Hampson (1975) were not included in the study. David Lester applied the same methodology to the contiguous 48 states of USA (Lester, 1996) and to the data sets of 18 and 32 nations (Lester, 2000). As concluded by Lester, factor analyses of the demographic indices from USA and 32 nations seemed to support Lynn’s (1971) and Lynn and Hampson’s (1975) findings. For example, suicide, accident and liver cirrhosis mortality seemed to reflect neuroticism whereas divorce and crime rates were related to extraversion. It should be noted, however, that some indicators (e.g. divorce rate, crime rate) loaded on different factors in the US and international data.
In the studies using 1960s data by Lynn (1971) and Lynn and Hampson (1975), identification of social indicators as indices of neuroticism or extraversion was based on theoretical interpretation of factor matrices, probably because EPQ scores were not available at that time. On the other hand, the main objective of the study by Lynn and Martin (1995) was to investigate the personality correlates of national work ethic and competitiveness, so only the relevant social indices were included. Hence, the full pattern of the relationships between neuroticism and extraversion measured by social indicators and EPQ N and E scales are not known yet.
The aim of the present study is to investigate the relationship between EPQ personality scores (neuroticism and extraversion) and the corresponding social indicators of personality.
