6. Statistics can be interpreted in many different ways after the information has been gathered. When such interpretations produce policies which are disadvantageous to women, something must be done. We need to be able to interpret statistics from several different points of view, in particular we need to be able to do this from the point of view of women.

Statistical interpretations tend to use universal quantifiers such as some, most, the majority, the most frequent, the most likely, average, etc. "Some" can mean anywhere from 1% to 99%, but not 0% nor 100%. "Most" or the "majority" means any number over 50%. "Most likely", as in most likely choice or most frequent choice, means that this choice was chosen more often than any other choice. However, note that the "most likely" choice could be chosen by as few as (eg.) 25%, if the next most likely choice happens to be 24%.

An average is generally an arithmetic mean value, obtained by adding all the values provided by all respondents and then dividing the total by the number of respondents. That is, it is obtained by a mathematical procedure and is not necessarily a value reported by any respondent. The sum of the values above the mean is equal to the sum below, although the number of respondents above the mean is not necessarily the same the number below.. This mean can be badly skewed off centre. For example, when calculating the participation rate in the labour force of women 15 years and over, who are divorced, separated, or widowed, the group includes a disproportionately large number of widows over 65 years. These older women have very low participation rates and the bulk of their numbers tends to lower the average participation rate for the entire group.

The median is another measure of central tendency. It is the middlemost value of all those reported. For example, if there are 35 individual values, the median is the 18th when all the values are placed in order. The mean and the median are not necessarily the same. In the median, the number of respondents is equal above and below the median, but the actual sum of the values themselves may be quite different. For example, in 1975 the annual earnings for female heads of households showed the following values:

In each case there are a much greater proportion of women earning small incomes (i.e. below the mean) than there are women earning large incomes. The median values indicate that the middlemost woman earns considerably less than the average value for all women. This discrepancy is greatest for women not in the labour force and least for women in the labour force.