Question

Beginning in the 1970s, anthropologists began to address the problem of androcentrism (overvaluing of male gender) within the field of anthropology. Since then, subtle gender biases have been found. Specifically, in a 1990 publication, Catherine Lutz found that in the journals American Ethnologist, American Anthropologist, and Ethos, male authors failed to reference publications by women as often as female authors did. In a 1985 publication, Joan Gero found that male archaeologists were more likely to do fieldwork dissertations while female archaeologists were more likely to do labwork dissertations. Specifically, between 1980 and 1984, 176 male dissertations in archaeology were fieldwork-oriented while 107 were lab-oriented. In contrast, 60 female dissertations in archaeology were fieldwork-oriented while 115 were lab-oriented. This led Gero to conclude that there was what she called a “woman at home” ideology, in which men were encouraged to go outside the confines of the university and bring home the data, while women were expected to stay within the university (to “stay home”) and process (“cook”) the data. Gero argued that this pattern in archaeology reflected broader expectations that the public sphere was gendered male while the domestic sphere was gendered female. Is the pattern in Gero’s data statistically significant? Also, use a strength test to calculate the strength of the association of men with fieldwork and women with labwork.

Answer 1

Let us summarize what is given in the problem. From the statements, we form a 2X2 contingency table as below:

	Field	Lab	Total
Male	176	107	283
Female	60	115	175
Total	236	222	458

From this data we need to assess

Gero argued that this pattern in archaeology reflected broader expectations that the public sphere was gendered male while the domestic sphere was gendered female. Is the pattern in Gero’s data statistically significant?

That is we need to assess whether the gender and the homing(Women at home) are associated.

Null Hypothesis:

The two factors gender and Homing are not associated or are independent

Alternative Hypothesis:

There is an association between two factors.

Test statistics:

We shall use the

Where is the Observed Frequency and is the observed frequency and

Where R is the row total of the cell and C is the column total for the cell, and T is the overall total.

For example in the above table, the for 176(Male field is) 283*236/458=145.825. We need to calculate for all the 4 entries and then calculate the Test statistic.

Expected Values

  

	Field	Lab	Total
Male	145.8253	137.1747	283
Female	90.1747	84.8253	175
Total	236	222	458

We then calculate the statistics as

The tabulated value for at (2-1)x(2-1) =1 df is 3.841 at 5% level of significance.

Since the calculated value is more than the tabulated value, the null hypothesis is rejected and hence we conclude that there is an association between the gender and the home.

Strength Test:

We shall use a Phi coefficient:

where a,b,c and d are the cell counts.

The interpretation is the same as the correlation coefficient. Here, it is 0.2713 which indicates a little or no association between these two factors.