URC

Estimating Suicide Rates in Nations that Do Not Report Suicide Statistics

Zorel Zambrano and Lawrence T. White*
Beloit College


Abstract

We report a new method to estimate suicide rates in nations that do not collect or report suicide statistics. Using indicators of suicide rates in a sample of 73 nations and standard regression techniques, we identified four predictors—divorce rate, locus of control, per capita GDP, and fertility rate—and generated different regression equations. These equations appear to produce reasonably valid estimates of national suicide rates.

Introduction

What is the suicide rate in Bolivia? Indonesia? Kenya or Taiwan? No one knows because government officials in these countries either do not collect or do not report official suicide statistics. Indeed, there are more than 100 nations that do not report suicide statistics to international organizations such as the World Health Organization.

This lack of reporting is unfortunate, as suicidal behavior is a public health problem in some countries (Platt, Bille-Brahe, Kerkhof, Schmidtke, Bjerke, Crepet et al., 1992), and statistics are useful to government planners, public health officials, medical researchers, and social scientists. When official statistics are unavailable, estimates can be used in their place. But how can we estimate suicide rates?

In this paper, we report a new method to estimate suicide rates in nations (or regions) that do not report official suicide statistics. The strategy is conceptually simple. First, identify indicators (e.g., divorce rate) that are correlated with national suicide rates and readily accessible for most countries. Second, use stepwise regression procedures to identify the best predictors of suicide rates and generate a prediction equation. Third, use the equation to estimate the suicide rate in a nation for which values are available for the predictors in the equation.

This exercise is a purely predictive one, with no implications for the causal status of indicators. We also acknowledge the imperfect nature of suicide statistics. Suicide is underreported, and official rates are subject to error. There is evidence, however, that these sources of error do not invalidate the differences reported between cultural and social groups (e.g., Sainsbury, 1983). Thus, we believe official suicide statistics are sufficiently reliable to justify their use by researchers to compare rates over time and between nations.

Known Indicators of National Suicide Rates

In this section, we list and briefly discuss 10 variables that have been identified as statistically significant predictors of national suicide rates.

  1. Numerous studies have found a statistical relationship between normative religious beliefs (as indicated by church attendance, church membership, or religious sanctions against suicide) and national or regional suicide rates (e.g., Huang, 1996; Kelleher, Chambers, Corcoran, Williamson, & Keeley, 1998; Neeleman, Halpern, Leon, & Lewis, 1997). Across different regions of the United States, higher levels of Catholic Church membership are associated with lower suicide rates (Burr, McCall, & Powell-Griner, 1994). The Ukraine’s western provinces, where more people attend church, have lower suicide rates than its eastern provinces, where fewer people attend church (Kondrichin & Lester, 2002). Nations that publish relatively more religious books tend to have lower suicide rates (Cutright & Fernquist, 2001; Fernquist, 2003a).
  2. Nations with relatively high divorce rates tend to have higher suicide rates (Fernquist, 2003a; Johnson, Krug, & Lloyd, 2000; Lester, 1997; Lester & Rihmer, 1992).
  3. Nations and regions with relatively high literacy rates tend to have higher suicide rates (Durkheim, 1952; Mayer, 2003), even when per capita GDP is controlled (Marusic, Khan, & Farmer, 2002).
  4. Nations that are relatively individualistic (i.e., emphasize values of independence and autonomy) tend to have higher suicide rates (Allik & Realo, 1997; Eckersley & Dear, 2002; Rudmin, Ferrada-Noli, & Skolbekken, 2003).
  5. Wealthier nations (as indicated by per capita GDP) tend to have higher suicide rates (Allik & Realo, 1997; Marusic, Khan, & Farmer, 2002).
  6. Nations with relatively high IQ scores tend to have higher suicide rates (Lester, 2003), even when per capita GDP is controlled (Voracek, 2004). There is some evidence this relationship also holds within nations (Voracek, 2005).
  7. Higher suicide rates may be associated with more extreme geographic latitudes. In the United States, northern states tend to have higher suicide rates than southern states (Lester & Shephard, 1998). In Argentina, southern regions have higher suicide rates than northern regions (Lawrynowicz & Baker, 2005). Within Europe, suicide rates generally increase with increasing latitude, although the strongest pattern is a J-shaped curve that points toward northeastern Europe, where suicide rates are higher (Voracek, Fisher, & Marusic, 2003).
  8. Nations with a relatively high percentage of women in the labor force tend to have higher suicide rates (Cutright & Fernquist, 2001).
  9. Nations with relatively high fertility rates tend to have lower suicide rates (Cutright & Fernquist, 2001).
  10. Nations with relatively high levels of external locus of control (LOC) tend to have higher suicide rates (Allik & Realo, 1997). LOC is a psychological construct that refers to a generalized belief in a person’s ability to control the things that happen in one’s life.

Still other variables have been found to correlate with suicide rates. For example, national suicide rates are statistically associated with cancer mortality rates (Bridges & Williamson, 2003), alcohol consumption (Lester, 1995; Lester, 1997; Lester, 2001), firearm availability (Johnson, Krug, & Potter, 2000), and perceived income inequality (Fernquist, 2003b). In theory, these variables could be used to estimate suicide rates in various nations, but the values of these variables are unavailable in many countries (e.g., African and Islamic) for which we wish to estimate suicide rates. This fact does not pose a problem, however, because it is not necessary to include all indicators in a regression formula. We need only include those that (a) are likely to account for a significant proportion of the variation in national suicide rates and (b) are available for those nations whose suicide rates we wish to estimate. We believe we have identified a sufficient number of indicators that meet these criteria.

Armed with these 10 indicators, we attempted to construct several regression equations that, taken together, would allow us to estimate national suicide rates with a fair degree of accuracy.

Method

Sample of Nations

Our sample consisted of 73 nations for which suicide data are available from the World Health Organization. Most of the nations in our sample are located in the Americas, Europe, Central Asia, and the Pacific. Relatively few nations are located in Africa, Southeast Asia, and the Middle East.

Variables

The variables included in this study were national suicide rate and the 10 indicators noted earlier. When possible, we calculated aggregated values for these 11 variables. Estimates based on aggregated data (e.g., multiple observations over time) are generally more accurate than estimates based on single observations (Epstein, 1983).

Using statistics from the World Health Organization (2005), we calculated the average suicide rate (i.e., number of suicides per year per 100,000 persons) for the years 1992-2003 for each of the 73 nations. Not all nations reported data for all years between 1992 and 2003. Cuba, for example, reported its suicide rates for 1992, 1995, and 1996. In these cases, we averaged all available values.

Using statistics from the United Nations (2003), we calculated the average per capita GDP (in U.S. dollars) for each nation. Per capita GDP refers to the total value of goods produced in a year divided by the total population of the country. To calculate the average per capita GDP for a nation, we used the same years for which suicide data were reported. For example, Cuba reported suicide rates for 1992, 1995, and 1996, so we used GDP values for 1992, 1995, and 1996 to calculate Cuba’s average per capita GDP.

Using statistics from the United Nations Statistics Division (2005), we recorded the divorce rate (i.e., number of divorces per year per 1,000 persons) for 58 nations. We calculated the average divorce rate for the years 1996-2000 for each nation.

Using statistics from the United Nations (2003), we recorded religious book production for 65 nations. We divided the number of religious books published by the total number of books published in a given year—and then averaged those rates across the years for which data were available.

Using statistics from the United Nations (2005a), we recorded women’s participation in the labor force for 70 nations. We recorded, for the most recent year available, the percentage of adult women who were engaged in economic activity. We also recorded the fertility rate for 73 nations. We recorded, for the most recent year available, the total number of children a female was predicted to bear, assuming she lives through her childbearing years.

Using data from Smith, Trompernaars, and Dugan (1995) and Allik and Realo (1997), we recorded average locus of control scores for 32 nations. Rotter’s Locus of Control (LOC) scale is a measure of generalized beliefs in one’s ability to control the things that happen in one’s life. On Rotter’s scale, a high score indicates a high degree of externality and a low score indicates a high degree of internality.

Using data from Hofstede (2005), we recorded individualism rankings for 34 nations. A high individualism rank indicates a society that places more emphasis on individual independence and autonomy. A low individualism rank indicates a society that values group (e.g., family) goals over individual goals.

Using data from Lynn and Vanhanen (2002, Table 6.5, pp. 73-80), we recorded estimates of national intelligence for 69 nations. National IQs for most nations were based on direct evidence (i.e., scores aggregated across studies that measured intelligence); national IQs for remaining nations were estimated from national IQs of neighboring or otherwise comparable nations.

Using statistics from the United Nations (2005b), we recorded the adult literacy rate for 73 nations. We recorded the percentage of males and females, aged 15-24, who can both read and write a short, simple statement about everyday life.

Finally, using information compiled by the Getty Trust (2005), we recorded each nation’s geographic latitude. We recorded the latitude (in degrees) of each nation’s capital city, as a nation’s capital is often the largest city.

In sum, we obtained values for national suicide rates and 10 different indicators. Fifty-five nations had missing values for one or more indicators. The remaining 18 nations are mostly in Western Europe and North America.

Results and Discussion

In our sample of 73 nations, seven of the 10 indicators were significantly correlated with national suicide rates (see Table 1). In every instance, the direction of the correlation was as expected, given the results of earlier studies.

Table 1
Zero-Order Correlations between National Suicide Rate and 10 Indicators

Indicators

Correlation with Suicide Rate

Divorce Rate

.652** (n = 58)

Locus of Control

.584** (n = 32)

Fertility Rate

-.572** (n = 73)

Individualism Rank

.564** (n = 34)

Distance from the Equator

.494** (n = 73)

National IQ

.388** (n = 69)

Adult Literacy Rate

.300** (n = 73)

Women’s Participation in the Labor Force

.088 (n = 70)

Gross Domestic Product (per capita)

.078 (n = 73)

Religious Book Production

-.042 (n = 65)

** Correlation is significant at the .01 level (2-tailed).
Note: Values in table are Pearson’s product-moment correlation coefficients (r) with one exception; the value for Individualism Rank is Spearman’s rho.

We were initially surprised by one finding in particular. Earlier studies (Allik & Realo, 1997; Marusic, Khan, & Farmer, 2002) have found a strong, positive correlation between per capita GDP and national suicide rates. In our sample, however, the relationship (r = .08) was weak and insignificant. We created a scatterplot to determine the exact nature of the relationship between national suicide rates and per capita GDP. The plot revealed two clusters of nations, 46 poorer nations (with per capita GDP under $10,000) and 27 wealthier nations (with per capita GDP over $10,000). Among the poorer nations, per capita GDP and national suicide rate were not correlated (r = .08, n.s.). Among wealthier nations, however, per capita GDP and national suicide rate were strongly correlated (r = .61, p < .01). These findings suggest that a positive relationship between a nation’s wealth and its suicide rate will be found only among relatively wealthy nations. Indeed, a re-examination of earlier studies revealed that their samples included wealthier, mostly European nations.

The weakest correlate in our sample was religious book production (r = -.04, n.s.). This seems counter-intuitive, although a close examination of national suicide rates reveals several religious countries with high suicide rates. Lithuania, for example, is a Catholic country that, according to the World Health Organization, can claim the world’s highest suicide rate in recent years. Nevertheless, we believe the presence of religious sanctions has a dampening effect on suicide rates. How then can we explain the weak correlation between religious book production and suicide rates?

Two possibilities come to mind. First, religious book production appears to be an unreliable indicator because the rate in a given country often fluctuates dramatically from year to year. This suggests that religious book production is affected substantially by extraneous variables that change over relatively short periods of time. Second, countries with religious sanctions against suicide are less likely to report their suicide rates to the World Health Organization (Kelleher, Chambers, Corcoran, Williamson, & Keeley, 1998). As a result, these nations were less likely to be included in our sample. (Indeed, most of the world’s Islamic nations are absent from our sample of 73 nations). The weak correlation between religious book production and suicide rates in the present study may be due to a restricted range problem.

We should note another interesting finding. Locus of control, which is strongly correlated with national suicide rate (r = .58, p < .01), was significantly correlated with only one other indicator (fertility rate, r = -.54, p < .01). This suggests that LOC—one of the few truly psychological variables in our analysis—will emerge as a significant predictor in a regression equation because its contribution will be relatively independent of the contributions made by other indicators, many of which are correlated with each other.

We conducted a series of stepwise regressions to construct three different prediction models. In the first regression, variables with missing values were excluded listwise, which resulted in a prediction model based on 18 nations only (see Table 2). The 18 nations were Australia, Brazil, Japan, Mexico, New Zealand, the United States, and 12 European countries. The significant predictors in this model (Model A) were per capita GDP and LOC, which together accounted for 64% of the variance. Model A’s prediction equation is Suicide Rate = -11.405 + GDP1000(.399) + LOC(1.69).

Table 2
Summary of First Stepwise Regression to Predict Suicide Rate (N = 18 nations)

Variable

B

SE B

β

Model 1

 

 

 

Per Capita GDP
(in units of 1000 USD)

.464

0.1

0.722

Model 2

 

 

 

Per Capita GDP
(in units of 1000 USD)

.399

0.99

0.620

Locus of Control

1.69

0.77

0.356

Note: R² = .521 for Model 1; R² = .638 for Model 2.

In the second regression, variables with missing values were excluded pairwise, which meant that all 73 nations contributed in some degree to the prediction model (see Table 3). The significant predictors in Model B were divorce rate and LOC, which together accounted for 79% of the variance. Model B’s prediction equation is Suicide Rate = -38.576 + DIV(6.384) + LOC(4.165).

Table 3
Summary of Second Stepwise Regression to Predict Suicide Rate (N = 73 nations)

Variable

B

SE B

β

Model 1

Divorce Rate

6.233

1.546

.652

Model 2

Divorce Rate

6.384

.963

.668

Locus of Control

4.165

.697

.602

Note: R² = .425 for Model 1; R² = .787 for Model 2.

In the third regression, we excluded two indicators—LOC and individualism—because their values are not readily available for most nations in the world. Variables with missing values were excluded listwise, which resulted in a prediction model based on 53 nations (see Table 4). All regions of the world, except Africa and the Middle East, were represented. The significant predictors in Model C were divorce rate, fertility rate, and GDP, which together accounted for 54% of the variance. Model C’s prediction equation is Suicide Rate = 17.272 + DIV(5.452) + FERT(-6.195) + GDP1000(-.158).

Table 4
Summary of Third Stepwise Regression to Predict Suicide Rate (N = 53 countries)

Variable

B

SE B

β

Model 1

Divorce Rate

6.089

1.006

.647

Model 2

Divorce Rate

4.845

1.023

.514

Fertility Rate

-6.329

2.109

-.326

Model 3

Divorce Rate

5.452

1.038

.579

Fertility Rate

-6.195

2.049

-.319

Per Capita GDP
(in units of 1000 USD)

-0.158

.079

-.204

Note: R² = .418 for Model 1; R² = .507 for Model 2; R² = .544 for Model 3.

Each prediction model has its own advantages and disadvantages. Model A cannot be used when information about Locus of Control is unknown (as is often the case) but may be most appropriate when predicting the suicide rate of a Western nation or region (e.g., southern Germany). Model B also cannot be used when LOC is unknown, but the model is based on statistical information from all 73 nations and boasts a multiple R 2 of 0.79. Model C accounts for less variation than the other models, but the values needed are readily available for many nations.

How accurate are the predictions made by these models? And are the predictions fairly similar or disparate? The first question refers to validity and the second refers to reliability (i.e., stability). We explored the issues of validity and reliability through an analysis of suicides in Turkey, an Islamic country whose government does not report suicide statistics to the World Health Organization. As a result, Turkey could not be included in our initial analyses. Fortunately, values of the four predictor variables are available for Turkey: GDP (in 1000 USD) = 2.831, LOC = 8.91, DIV = 0.50 and FERT = 2.43. We entered these values into the equations and produced three estimates of the suicide rate in Turkey. The estimates were 4.78, 1.73, and 4.50 suicides per year per 100,000 persons, for models A, B, and C respectively.

The estimates are fairly similar to each other, given the large variability in national suicide rates. In our sample of 73 nations, aggregated suicide rates ranged from a low of .75 (Azerbaijan) to a high of 44.85 (Lithuania), with a mean of 13.26 and standard deviation of 9.96. The difference between our highest and lowest estimates of Turkey’s suicide rate was 3.05, or less than one-third of the standard deviation observed in national suicide rates.

Are our estimates of Turkey’s suicide rate reasonably accurate? It is impossible to know with a high degree of certainty, as even official suicide statistics are estimates of a sort. Fortunately, policymakers may not need precise estimates of suicide rates; it may be enough to know if the rate is unusually low, relatively low, moderate, relatively high, or unusually high. Suicide prevention programs are not established (or abolished) because the suicide rate has increased (or decreased) a few percentage points; such programs are established (or abolished) when policymakers believe the suicide rate is unacceptably high (or tolerably low). In short, there appears to be little need for precise estimates of suicide rates; reasonably accurate estimates will suffice.

We believe our estimates of Turkey’s suicide rate are reasonably accurate. According to Devrimci-Ozguven and Sayil (2003), Turkey’s State Institute of Statistics collects and publishes suicide statistics. They report that the completed suicide rate in Ankara (Turkey’s second largest city) in 1998 was “5.5 per 100,000, including women and men of all ages” (p. 325). They also note that Ankara’s suicide rate is probably higher than the national rate, which likely would place Turkey’s overall suicide rate within the range of our estimates (i.e., 1.73 – 4.78).

One might argue that prediction models are not needed to estimate the suicide rate in countries such as Turkey because “everyone knows” that poor, Muslim countries have low suicide rates and rich, Protestant countries have high suicide rates. The point is a legitimate one. For our prediction models to be useful, they should be able to generate reasonably accurate estimates for nations whose suicide rates cannot easily be approximated on the basis of wealth, religion, or location.

One such nation is Australia. Social scientists likely would predict Australia’s suicide rate to be above average, given its relative wealth and low religiosity. As we will see, the social scientists’ prediction would be substantially off the mark.

As a further test of the utility of our models, we estimated the suicide rate in Australia in 2003. Fortunately, values of the four predictor variables are available for Australia in that year: GDP (in 1000 USD) = 30.7, LOC = 8.23, DIV = 2.6 and FERT = 1.76. When we entered these values into the prediction equations, we produced three estimates of Australia’s suicide rate: 14.75, 12.30, and 15.69 suicides per year per 100,000 persons, for models A, B, and C respectively.

These estimates are close approximations to Australia’s actual suicide rate in 2003. The World Health Organization does not report Australia’s 2003 suicide rate, but we were able to calculate the number by searching Australian newspapers on-line for government news about suicide statistics. According to several sources, 2213 persons committed suicide in Australia in 2003. Given Australia’s 2003 population of 19.86 million, we calculated a rate of 11.14 suicides per 100,000 persons. This number is remarkably close to Model B’s estimate of 12.30. (Recall that Model B is “better” than Models A and C because it accounts for a higher percentage of the variance—79%.) In short, our prediction equations appear to generate reasonably accurate estimates, even for nations like Australia whose suicide rates are not easily “guesstimated.”

In conclusion, suicidal behavior is a public health problem in many countries. As a result, government and public health officials are eager to know how many suicides occur in their country or region each year. Unfortunately, officials in many countries do not know the suicide rate because official statistics are not collected or reported. When knowing is not possible, estimating is the next best thing. Using standard stepwise regression techniques, we identified four predictors of national suicide rates—divorce rate, locus of control, per capita GDP, and fertility rate—and generated three different prediction formulas, each with its own advantages and disadvantages. Taken together, they appear to provide reasonably accurate estimates of suicide rates in nations that do not report official suicide statistics. The knowledge generated will be imperfect knowledge, but imperfect knowledge is sometimes “good enough.”


References

Allik, J., & Realo, A. (1997). Psychological and cultural mechanisms of suicide. TRAMES, 4, 306-321.

Bridges, F. S., & Williamson, C. B. (2003). Association of national cancer mortality and suicide rates. Perceptual and Motor Skills, 97(2), 424-426.

Burr, J. A., McCall, P. L., & Powell-Griner, E. (1994). Catholic religion and suicide: The mediating effect of divorce. Social Science Quarterly, 75(2), 300-318.

Cutright, P., & Fernquist, R. (2001). The age structure of male suicide rates: Measurement and analysis of 20 developed countries, 1955-1994. Social Science Research, 30, 627-640.

Devrimci-Ozguven, H., & Sayil, I. (2003). Suicide attempts in Turkey: Results of the WHO-EURO Multicentre Study on Suicidal Behavior. Canadian Journal of Psychiatry, 48(5), 324-329.

Durkheim, E. (1952). Suicide: A study in sociology. Translated by J. A. Spaulding & G. Simpson. London: Routledge and Kegan Paul

Eckersley, R., & Dear, K. (2002). Cultural correlates of youth suicide. Social Science & Medicine, 55(11), 1891-1904.

Epstein, S. (1983). Aggregation and beyond: Some basic issues in the prediction of behavior. Journal of Personality, 51, 360-391.

Fernquist, R. M. (2003a). Does the level of divorce or religiosity make a difference? Cross-national suicide rates in 21 developed countries. Archives of Suicide Research, 7(3), 265-277.

Fernquist, R. M. (2003b). Perceived income equality and suicide rates in central/eastern European countries and western countries, 1990-1993. Death Studies, 27(1), 63-80.

Getty Trust. (2005). Getty thesaurus of geographic names. Retrieved on June 15, 2005, from http://www.getty.edu/research/conducting_research/vocabularies/tgn/index.html.

Hofstede, G. (2005). Cultural dimensions. Retrieved on June 15, 2005, from http://www.geert-hofstede.com/hofstede_dimensions.php.

Huang, W-C. (1996). Religion, culture, economic and sociological correlates of suicide rates: A cross-national analysis. Applied Economics Letters, 3(12), 779-782.

Johnson, G., Krug, E., & Potter, L. (2000). Suicide among adolescents and young adults: A cross-national comparison of 34 countries. Suicide and Life-Threatening Behavior, 30(1), 74-82.

Kelleher, M. J., Chambers, D., Corcoran, P., Williamson, E., & Keeley, H. S. (1998). Religious sanctions and rates of suicide worldwide. Crisis: The Journal of Crisis Intervention and Suicide Prevention, 19(2), 78-86.

Kondrichin, S. V., & Lester, D. (2002). Suicide in the Ukraine. Crisis: The Journal of Crisis Intervention and Suicide Prevention, 23(1), 32-33.

Lawrynowicz, A. E. B., & Baker, T. D. (2005). Suicide and latitude in Argentina: Durkheim upside-down. American Journal of Psychiatry, 162(5), 1022.

Lester, D. (1995). The association between alcohol consumption and suicide and homicide rates: A study of 13 nations. Alcohol and Alcoholism, 30(4), 465-468.

Lester, D. (1997). Suicide in an international perspective. Suicide and Life-Threatening Behavior. Special Issue: Individual, Cultural, International Perspectives, 27(1), 104-111.

Lester, D. (2001). Association of alcohol use and suicide in 27 nations of the world. Psychological Reports, 88(3, Part 2), 1129.

Lester, D. (2003). National estimates of IQ and suicide and homicide rates. Perceptual and Motor Skills, 97(1), 206.

Lester, D., & Rihmer, Z. (1992). Sociodemographic correlates of suicide rates in Hungary and the United States. Psychiatria Hungarica, 7(5), 453-457.

Lester, D., & Shephard, R. (1998). Variation of suicide and homicide rates by longitude and latitude. Perceptual and Motor Skills, 87(1), 186.

Lynn, R., & Vanhanen, T. (2002). IQ and the wealth of nations. Westport, CT: Praeger.

Marusic, A., Khan, M., & Farmer, A. (2002). Can the interaction of poverty and literacy explain some variability of suicide rates in Europe? European Journal of Psychiatry, 16(2), 103-107.

Mayer, P. (2003). Female equality and suicide in the Indian states. Psychological Reports, 92, 1022-1028.

Neeleman, J., Halpern, D., Leon, D., & Lewis, G. (1997). Tolerance of suicide, religion and suicide rates: An ecological and individual study in 19 Western countries. Psychological Medicine, 27, 1165-1171.

Platt, S., Bille-Brahe, U., Kerkhof, A., Schmidtke, A., Bjerke, T. Crepet, P., and others. (1992). Parasuicide in Europe: The WHO/EURO Multicentre Study on Parasuicide. I. Introduction and preliminary analysis for 1989. Acta Psychiatrica Scandinavica, 85, 97-104.

Rudmin, F. W., Ferrada-Noli, M., & Skolbekken, J . (2003). Questions of culture, age and gender in the epidemiology of suicide. Scandinavian Journal of Psychology, 44(4), 373-381.

Sainsbury, P. (1983). Validity and reliability of trends in suicide statistics. World Health Statistics Quarterly, 36(3-4), 339-348.

Smith, P. B., Trompenaars, F., & Dugan, S. (1995). The Rotter locus of control scale in 43 countries: A test of cultural relativity. International Journal of Psychology, 30(3), 377-400.

United Nations Statistics Division. (2005). Demographic yearbook 2000. Retrieved on June 15, 2005, from http://unstats.un.org/unsd/demographic/products/dyb/dyb2000.htm.

United Nations. (2003). United Nations Statistical Yearbook 2000. New York: United Nations Publications.

United Nations. (2005a). Statistics and indicators on women and men. Retrieved on June 15, 2005, from http://unstats.un.org/unsd/demographic/products/indwm/ww2005/tab5d.htm.

United Nations. (2005b). Millennium development goal indicators database. http://millenniumindicators.un.org/unsd/mi/mi_series_results.asp?rowID=656.

Voracek, M. (2004). National intelligence and suicide rate: An ecological study of 85 countries. Personality and Individual Differences, 37, 543-553.

Voracek, M. (2005). The social ecology of intelligence and suicide in Belarus. Journal of Social Psychology, 145, 613-617.

Voracek, M., Fisher, M. L., & Marusic, A. (2003). The Finno-Ugrian suicide hypothesis: Variation in European suicide rates by latitude and longitude. Perceptual and Motor Skills, 97(2), 401-406.

World Health Organization. (2005). Suicide prevention country reports. Retrieved on June 15, 2005, from http://www.who.int/mental_health/prevention/suicide/country_reports/en/.


Author Note

This research was supported by stipends provided to the authors by the Ronald E. McNair Postbaccalaureate Achievement Program. We thank Jüri Allik and Charles Westerberg for their helpful comments.

Correspondence concerning this article should be addressed to Lawrence T. White, Department of Psychology, Beloit College, 700 College Street, Beloit, Wisconsin 53511. E-mail: WhiteLT@beloit.edu.

 

 


URC RESOURCES:

©2002-2016 All rights reserved by the Undergraduate Research Community.

Research Journal: Vol. 1 Vol. 2 Vol. 3 Vol. 4 Vol. 5 Vol. 6 Vol. 7 Vol. 8 Vol. 9 Vol. 10 Vol. 11 Vol. 12 Vol. 13 Vol. 14 Vol. 15
High School Edition

Call for Papers ¦ URC Home ¦ Kappa Omicron Nu

KONbutton K O N KONbutton