National Differences in Personality and Predictability of Gender From Personal Names

Abstract

This article reports a test of the hypothesis that national differences in personality traits are expressed by national differences in how accurately the final letter of the personal name designates the female or male gender. The names were obtained from lists of more than 80 popular names in each nation, separately for females and males. Gender designation was more accurate for female than male names. In nations with more accurate gender designation of the final letter of first names, four personality traits self-reported more often by inhabitants are high uncertainty avoidance (UA), high power distance (PD), low individualism (Ind), thereby high collectivism, and a low proportion who are very happy (VH).

Keywords

personal names gender uncertainty avoidance power distance (PD)individualism (Ind)

A cultural variable in contemporary nations is the accuracy with which the personal name designates the female or male gender. In the Latin language, dominant in the ancient Roman Empire, singular nouns in the first declension are usually feminine and end in a. Singular nouns in the second declension are always masculine and end in us. A personal name is a type of noun. According to Ingraham (1997), in the Roman Empire most personal names given to females and males were the same except that the ending always was a for females, us for males. Examples are Claudia and Claudius, Tertulia and Tertulius.

During the span of more than 1,500 years after the Roman Empire was conquered by Germanic tribes, multiple nations developed with different dominant languages. The first and second declensions disappeared or became less rigid. Nevertheless, different final letters of personal names continue to be associated with the female or male gender of personal names. The present article reports a test of the hypothesis that national differences in personality traits are expressed by national differences in how accurately the final letter of the personal name designates the female or male gender.

A book by Teresa Norman (1996) identifies recent choices of more than 80 popular first names, separately for females and males, in each of 14 contemporary nations. Contrary to the personal names of the ancient Romans, the accuracy of gender designation from the final letter of the first name is less than 100% in each of these contemporary nations. Differences among the nations enable us to explore whether the degree of accuracy of gender designation may be correlated with national differences in personality.

National differences in four general personality traits were initially defined by Hofstede (1980) and have been used by many other authors. Hofstede (2001) reported quantitative scores on each trait applied to a large sample of countries. One of the general traits is uncertainty avoidance. Our first hypothesis is that more accurate designation of female or male gender from the final letter of the first name will occur in nations whose residents have a high average score on uncertainty avoidance. Choices of names with high or low designation of gender can evolve in each nation during many generations of babies given a personal name.

A further hypothesis is that national traits highly correlated with uncertainty avoidance (UA) will be correlated with national accuracy of gender designation from the final letter of the first name. Accordingly, we identified three traits that are correlated with high UA in the cross-national literature: high power distance (PD), low individualism (Ind), thereby high collectivism (Hofstede, 2001), and a low proportion of inhabitants who describe themselves as very happy (VH; Minkov, 2009). In the results section, we compare the correlations in larger worldwide samples with the smaller sample used here.

Method

Information on each of 14 contemporary nations is shown in Table 1. The number of names (NN), the most frequent final letter (FL), the percentage of names with the most frequent final letter (PN), and the gender designation score (GD) are listed separately for female names and male names.

Table 1.

Separately for Females and Males, NN and the FL are Followed by PN With the FL. The GD Subtracts From PN the Percentage Names With the Same FL for the Opposite Gender. The Last Four Columns Show the Scores on UA, PD, Ind, and Being VH.

	Female names				Male names
	NN	FL	PN	GD	NN	FL	PN	GD	UA	PD	Ind	VH
Poland	218	a	91.7	91.0	270	n	17.4	16.5	93	68	80	−49
Russia	108	a	95.4	88.5	116	i	24.1	23.2	95	95	39	−138
Italy	246	a	90.7	88.0	219	o	80.4	80.0	75	50	76	−59
Czech Republic	105	a	90.5	87.3	124	v	20.2	20.2	74	57	58	−102
Spain	301	a	88.0	85.0	293	o	66.2	65.9	86	57	51	−74
Portugal	86	a	78.2	77.4	123	o	70.7	66.0	104	63	27	−52
Hungary	84	a	72.6	66.8	120	n	13.3	10.0	82	46	80	−60
Greece	263	a	60.5	60.4	179	s	67.0	62.4	112	80	35	−44
Germany	263	a	53.2	53.2	188	n	17.6	14.6	65	35	67	−45
France	257	e	84.4	52.1	226	n	20.5	19.3	86	68	71	+63
India	430	a	64.7	45.2	376	n	12.5	11.6	40	77	48	+11
United Kingdom	651	a	38.7	38.3	448	n	22.8	16.2	35	35	89	+123
Ireland	323	a	35.3	32.3	469	n	30.5	17.8	35	28	70	+124
United States	468	a	31.2	30.1	442	n	20.8	10.5	46	40	91	+89

Note: NN = number of names. FL = most frequent final letter. PN = percentage names. GD = gender designation score. UA = uncertainty avoidance. PD = power distance. Ind = individulaism. VH = very happy.

The dominant language family is Indo-European in 13 of the 14 nations. The exception is Hungary. Four Romance languages, directly derived from Latin, are dominant in Italy, Spain, Portugal, and France. Three nations (United Kingdom, Ireland, and United States) share the same dominant language. The ancient Latin language has influenced the naming customs in each of the 14 contemporary nations. The final letter of the name therefore is probably the principal determinant of its female or male gender in all or most of these nations.

An apparent purpose of the lists of popular names in the book by Norman (1996) is to help writers to select the names of fictional characters in each nation. The names included in the present analysis are limited to the primary names, printed in bold type. The tabulation omits variants, shortened names, and pet names.

The PN is shown for each nation in Table 1. France is the only nation in which the same final letter, e, is the most frequent for female names (84.4%) and also for male names (32.3%). The second FL for male names, n (20.5%), is substituted.

In Table 1, GD subtracts from the PN the percentage of names with the same final letter for the opposite gender. For example, in Poland GD of 91.0% for female names subtracts the 0.7% of male names with the final letter of a, not shown in the table, from the 91.7% of female names with the final letter of a. The GD for both genders, not shown in Table 1, is the average of the separate scores for females and males. In Poland, the score for both genders, 54.2%, is the average of the scores of 91.0% for female names and 17.4% for male names, shown in Table 1.

Hofstede (2001, pp. 500-502) contains scores for each of the 14 nations on high UA (Table 1) and on two other general traits, high PD (Table 1), and low Ind (Table 1), thereby high collectivism. The information was obtained from questionnaires given to multiple employees of the IBM Corporation in each nation.

UA (Table 1) was measured for employees in the work place. A high score is obtained from three criteria as follows: (a) Strong rule orientation, indicated by “yes” to the statement “Company rules should not be broken—even when the employee thinks it is in the company’s best interest;” (b) Employment stability, indicated by stated intention to continue with the company for two or more years; and (c) High stress, expressed in the answer to the question “How often do you feel nervous or tense at work?”

PD (Table 1) measures inequality in boss–subordinate relationships. Components of high PD are subordinates who are afraid to disagree with superiors, authoritarian decision-making by superiors, and subordinates preferring authoritarian decision-making by superiors.

Ind instead of collectivism (Table 1) is a general measure of social behavior. Individualistic people are expected to look after themselves and their immediate families. Collectivistic people are integrated into strong, cohesive groups, which protect them in exchange for unquestioning loyalty.

Scores on percentage of people who reported being VH (Minkov, 2009) were obtained from the World Values Survey Association (2008). Random probability samples of inhabitants were used where possible.

All of the 14 nations with adequate information in the book by Norman (1996) are included. Data analyses used SPSS, a statistical package for personal computers (Norusis, 2009). SPSS now is a component of the IBM Corporation.

Results

Before we undertook the tests of the hypotheses, we first determined, in the sample of 14 nations, the correlations of high UA with the three other traits. The correlation was .54, p = .014 with PD, .–64, p = .004 with Ind rather than collectivism, and –.72, p = .004 with a high proportion of people reporting themselves as being VH.

The 14 nations are a small proportion of the entire sample of 68 countries in Hofstede (2001) and 97 countries in Minkov (2009). There are 62 countries included in both Hofstede and Minkov. For all 68 countries, the correlation of high UA was .27, p = .026 with PD and –.23, p = .065 with Ind instead of collectivism. For all 62 countries the correlation of high UA with a high proportion of VH inhabitants was –.24, p = .058. It is evident that the 14 nations selected by Norman (1996) were not randomly selected from the entire samples selected by Hofstede and by Minkov.

In Table 1, the NN exceeds 80 for both genders in each nation. The FL for female names is a in each nation except France. The FL for male names is n in 8 nations and is distributed among four other final letters in 6 nations.

The sequence of 14 nations in Table 1 begins with the highest GD for female names in Poland and ends with the lowest score in the United States. The gender designations are higher for female than male names in each nation except Greece.

The information on the final letter of the personal name, separately for female and male names, is followed by the score for each of the 14 nations on UA, PD, Ind, and being VH.

The separate gender designation scores for female and male names have a positive correlation of .48 (p = .081). The scores in the same nation therefore tend to be relatively high or low for both genders instead of high for one gender and low for the other gender.

Table 2 shows the correlations of the four personality traits with quantitative measures of accurate female or male gender designation from the final letter of the personal name. Separate correlations are applied to female names, male names, and the average for both genders.

Table 2.

In 14 Nations, Four Personality Traits are Correlated With Accurate Gender Designation From the Final Letter of the Personal Name. Separate Correlations are for Female Names, Male Names, and the Average of Both Genders.

	Correlation
Personality trait	Female	Male	Both
High UA	.73**	.50	.72**
High PD	.55*	.11	.36
Ind	−.49	−45	−.54*
VH	−.88***	−.37	−.71**

Note: UA = uncertainty avoidance. PD = power distance. Ind = individulaism. VH = very happy.

p < .05. **p < .01. ***p < .001.

The major hypothesis was that high UA would be associated with accurate gender designation of names. All three correlations with UA in Table 2 are positive. The correlations with female names and with the average of both genders are significant and strong. The correlations support our hypothesis that a nation whose inhabitants have high average UA usually contains personal names with a final letter that more accurately designates the female or male gender.

The three other variables related to high UA have correlations with accurate gender designation that are in the predicted direction: positive for high PD, negative for Ind, and for being VH. The correlation with gender designation of female names is even farther from zero for being VH (–.88) than for UA (.73).

A consistent difference between female and male names is portrayed in Table 2. All four personality traits have a correlation with accurate gender designation that is closer to zero for male names than for female names.

Discussion

The correlations of UA with the other three personality traits were higher for the sample of 14 nations than for the much larger samples of 68 countries (Hofstede, 2001) and 62 countries (Minkov, 2009). Most of the 14 nations selected by Norman (1996) have atypically large populations and active international commerce and travel. The findings might not apply to the larger samples of more diverse countries. The more homogeneous sample of 14 nations, including the United States, may be preferable because they vary adequately in accuracy of gender prediction and in the four personality traits.

According to Pinker (2011), the ancient Roman Empire was governed by very authoritarian and brutal rulers. According to Ingraham (1997), female and male names were perfectly designated from the final letter. We might infer that most of the inhabitants had high UA, high PD, low Ind, thereby high collectivism and were not VH. Many of the inhabitants were slaves, and some men were gladiators. Pinker (2011) also demonstrates that contemporary nations are more benevolent and less punitive. All of the 14 nations in our sample have less than 100% designation of female or male gender from the final letter of first names.

Several disadvantages limit the conclusiveness of the evidence that four personality traits are correlated with gender designation from the final letter of personal names. A single author (Norman, 1996) selected the sample of 14 nations and the popular names in each nation. Differences among nations in only four personality traits have been correlated with differences in accuracy of gender designation from the final letter of the first name. Three of the traits are from the same source (Hofstede, 2001), obtained from self-reports by employees of the IBM Corporation. The fourth trait, self-report of feeling VH (Minkov, 2009), is the only one with the advantage of being obtained from a probability sample of the population (World Values Survey Association, 2009). All four traits are determined by self-reports instead of by objective measurements.

Although the information has deficiencies, it tests and supports the hypothesis that in nations where the final letter of first names more accurately predicts the female or male gender of individuals, the inhabitants more often have the personality trait of UA and other traits highly correlated with UA. The measure of names and the four personality traits have the advantage of being choices by individuals. The information on names and on personality traits was obtained shortly before or after the end of the 20th century.

The most convincing difference obtained from the sample of 14 contemporary nations is that in comparison with male names, the FL of female names is more predominant over other final letters and is more highly correlated with the four personality traits. We speculate that many adults use different criteria for naming females and males. In the United States, Rossi (1965) found that females are more often given a name that is currently popular or that is perceived as being attractive. Males are more often given the name of the father, grandfather, or other relative, or else a traditionally popular name. Male names therefore change less over time and are less often chosen for the purpose of accurate gender designation.

If popularity is one of the reasons for choosing a girl’s name, then choosing a common ending may help explain both why female names have a higher proportion of the same ending letter and why doing more of it is associated with UA and the other personality traits. Another reason for the superior gender prediction of female names may be a gender difference in socialization. In a sample of mostly preliterate societies, girls are trained more strongly for obedience and responsibility, while boys are trained more strongly for self-reliance and achievement (Barry, Bacon, & Child, 1957; Barry, Child, & Bacon, 1959). Females are more often expected to conform to cultural customs, which can include a final letter of the first name that accurately designates the gender. More than 1,500 years after the Roman Empire ended, the FL of female names continues to be a in 13 contemporary nations; the exception, e, is France. The FL of male names continues to be s only in Greece.

Future research can and should expand our preliminary report on the correlation of four personality traits with accuracy of gender designation from the final letter of first names. A larger sample of contemporary nations is highly desirable. More numerous and diverse personality traits can be correlated with accuracy of gender prediction from the final letter of the first name. National differences in other attributes of first names might be correlated with national differences in personality traits.

Preferable to the book by Norman (1996) would be information on the most frequent personal names given in a single year if equivalent lists were available for the selected nations. Each new birth is an opportunity to continue or to change the type of name that is chosen. Changes in choices of names during a span of years could be compared with changes in personality traits of the inhabitants during the same span of years.

In the United States, recent increases in diversity of personal names probably are accompanied by less accurate gender designation. Barry and Harper (2004) found much greater diversity in choices of personal names for females and males born in 1990-1999 than in any previous decades recorded, beginning in 1900-1909 (Table 13, p. 135). Barry and Harper (2010) found that in the state of Pennsylvania, United States, diversity of personal names increased from 1990 to 2000. A potentially testable inference is that the increase in diversity of names was accompanied by less UA and by other traits associated with UA.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Herbert Barry III, is a professor emeritus at the University of Pittsburgh. His PhD in Experimental Psychology at Yale, 1957, led to behavioral psychopharmacology research using laboratory rats. Other topics of his diverse publications include cross-cultural comparisons, names given to female and male babies (coauthored with Aylene S. Harper), alcoholism, birth order in the childhood family, longevity of humans, Presidents of the United States, and optimal source of government revenue.

Aylene S. Harper, Ph.D. at University of Pittsburgh, 1975, is a professor of psychology in the Social Science Department at Community College of Allegheny County, South Campus. She also does psychological evaluations with assessment and therapy at both inpatient and outpatient health related settings.

References

Barry

Bacon

M. K.

Child

I. L.

(1957). A cross-cultural survey of some sex differences in socialization. Journal of Abnormal and Social Psychology, 55, 327-332.

Barry

Child

I. L.

Bacon

M. K.

(1959). Relation of child training to subsistence economy. American Anthropologist, 61, 51-63.

Barry

Harper

A. S.

(2004). The majority of female and male first names ended in A or E throughout the twentieth century. In Lee

J. W.

(Ed.), Gender roles (pp. 117-143). Hauppauge, NY: Nova Science Publishers.

Barry

Harper

A. S.

(2010). Racial and gender differences in diversity of first names. Names, 58, 47-54.

Hofstede

(1980) Culture’s consequences: International differences in work-related values. Beverly Hills, CA: SAGE.

Hofstede

(2001) Culture’s consequences: Comparing values, behaviors, institutions, and organizations across nations. Thousand Oaks, CA: SAGE.

Ingraham

(1997) People’s names: A cross-cultural reference guide to the proper use of over 40,000 personal and familial names in over 100 cultures. Jefferson, NC: McFarland.

Minkov

(2009) Predictors of differences in subjective well-being across 97 nations. Cross-Cultural Research, 43, 152-179.

Norman

(1996) A world of baby names. New York, NY: Penguin Putnam.

10.

Norusis

M. I.

(2009). PASW Statistics 18 statstical procedures companion. Upper Saddle River, NJ: Prentice Hall.

11.

Pinker

(2011) The better angels of our nature: Why violence has declined. New York, NY: Penguin Group.

12.

Rossi

A. S.

(1965) Naming children in middle-class families. American Sociological Review, 30, 499-530

13.

World Values Survey Association. (2008). World values survey 2005 official data file (v.20081015). Madrid, Spain: ASEP/JDS. Retrieved from http://www.worldvaluessurvey.com