Women-held Positions in World Governments by Dana Williams


Gender equality is often judged by the degree of democracy. The extent of democracy is often judged by the level of development that a particular democracy enjoys. With the issue of gender and democracy, it is telling to look at the involvement of women in government across the world. It stands to reason that the greater the proportion of governmental positions held by women, the greater extent of gender equality and thus democracy of that government.

True power in governments can sometimes be translated by the proportion of positions held, but also by the level of those positions. Thus, whereas the United States has roughly 33% of it's ministerial and sub-ministerial positions filled by women, the vast majority of higher-level positions are mainly male-dominated.1 Thus, although judgment can be rendered by the proportion of governmental offices filled by women to infer power and equality, true power in representative democracies that have high-levels of hierarchy require those top ranking positions to be filled by both genders for such equality and power to be shared.

Thus, the question arises: how do various national governments correlate with governmental participation of women and their status as developed or developing countries? Do significant differences exist in this proportion between these two classifications of nation states? And if so, does that reinforce the notion that developed countries are more (although not completely, since the United States has nearly the highest level of female participation, a percentage which is still only one-third of all governmental positions) gender-equitable and thus more democratic?

Problem and data review:
To answer that initial question, data has been collected from two secondary sources in the United Nations. The first classifies each nation state in the world as either "developed" or "developing". Sometimes these "developing" states are referred to as "un-developed", "3rd World", or "the South". These classifications, drawn from the UN Conference on Trade and Development, and while sensible in many respects, does question the motivations for categorizing some of the members of these classes. For instance, Greece and Portugal are often ranked much lower in a number of social indicators than Cuba and Russia, yet the former fall into the "developed" category and the latter into the "developing". The assertion that Russia, once considered a world super-power, is a "developing" country displays a few economic biases of UNCTAD's classification scheme. That aside, the vast majority of the classifications would be widely agreed upon.

The second set of data was retrieved from the Division for the Advancement of Women in the United Nations. They evaluated the percentage of women in government from all the world's nation states on two separate tiers: ministerial and sub-ministerial. In this study, the total of these two proportions was used for analysis. Things to consider with this data are that two separate factors are not evaluated or considered in a mere proportion. One, the size of this government and the number of governmental positions is not necessarily known. For instance, Andorra is the only state that ranks higher than the US is percentage at 38.5%, however the country itself is the size of the average US metropolitan area, which thus makes Andorra's figure not as revealing.2 The second factor is that the percentage of the population of a state that is female is not known. Certain states may not have completely balanced populations (for example the US is 51% female and 49% male), which can skew the interpretation of these statistics. For instance, in the US there is a slightly greater statistical likelihood of a woman having a government position than a male, all things being equal.

Yet all things are not equal, which is what this study hopes to determine. The states of the world have been divided up into either "developed" or "developing" classes, and are subsequently evaluated using a two-sample difference of proportions t-test.

In this study, 187 countries were tested. They include countries as large as China, India, and the US all the way to small countries such as Monaco, St. Lucia, and the Solomon Islands. When classified by the UNCTAD definition, 28 states were classified as "developed" and 159 as "developing". While this is obviously not a balanced classification, it does clearly gather the so-called "1st World" and "3rd World" together, at least by a Euro-centric definition. The mean percentage of female in government in the developed countries is 15%, while the mean percentage of the developing countries is half-8%.

The null hypothesis assumes that there is equality between both the developed and developing states in regards to women in government, while the alternate hypothesis asserts that there is a difference between those two classifications of states.

Since this problem has two samples (developed or developing states) and are independent of each other, a two-sample difference of means t-test is a useful test to use in order to determine if there is a difference between the percentage of women in government. In conducting this test, the software package SPSS was used, which also allows for computation of the data in a ranked format (without modification of the data type), such as the Wilcoxon W and Mann-Whitney U tests.

Results and conclusions:
Equal population variances are not assumed in this scenario, thus a separate variance estimate is used. The t test result for this problem ends up being 3.598 with a p-value (or two-tailed significance level) of .001. Thus, it can be said with 99.9% confidence that the null hypothesis can be rejected, thus leaving the alternate hypothesis that there is a definite difference between these two populations. In the use of a non-parametric test, such as the Wilcoxon W test, the resulting significance is computed at around .000, meaning that there is nearly complete confidence in rejection of the null hypothesis.

Thus, it may be asserted that there are differences in the percentage of women in governmental positions in developed and developing states. However, there is no indication of which of the two classification has a higher percentage of women in government since only a two-tailed test was used, which does not tell the direction of the significance (although it would be a relatively easy and safe guess that the developed states have a higher percentage than developing states).

The conclusion that one may draw from this study is that there is an explicit difference between developed and developing states in regards to women with governmental positions. If a two-tailed test were to be run, it would likely be further indicated that the developed states have higher percentages than developing states. If such results were discovered in a two-tailed test, then it would lead to the assumption that one potential way of raising up the percentage of women in government may be to further develop that country. Also, it would also be valid to say that raising the level of development in a country could also increase the percentage of women involved in government.

Future research that is possible with this notion of female participation in government was suggested by review of the data per country: it would be of interest to analyze the percentage of women participation in government by region of the world, such as Western Europe, Central America, Sub-Saharan Africa, etc. It is possible that similarities exist within these regions and could be brought out by an analysis of variance test done on sample data taken by world region.

# Division for the Advancement of Women, United Nations. (1996) Percent of Women In Government http://www.un.org/womenwatch/daw/public/percent.htm
# United Nations Conference on Trade and Development. (2000) UNCTAD Handbook of Statistics. http://www.unctad.org/en/docs/ps1tdstat25_tab1.en.pdf

1. Women percentages in highest-level government positions in the US are as follows: House of Representatives: 14%, Senate: 9%, President total: 0%.
2. Also interesting thing in the case of Andorra is the fact that although females hold nearly 40% of the total governmental positions, this number is obscenely enhanced by the sub-ministerial positions (75%) and not the primary ministerial positions (22%). This is a lucid example of how even a relatively high total percentage is thrown off by lower-level positions without truly affecting the upper-echelons of government.


Spatial Analysis