Online Adult Entertainment: Are Conservatives Really Buying More Porn Than Others?

As a lifelong student of the wider American culture and statistical correlations, I recently read with interest and now challenge, in part, the findings of a Journal of Economic Perspectives (Winter 2009) publication provokingly titled "Red Light States: Who Buys Adult Entertainment?" a study by assistant professor, Benjamin Edelman, of Harvard Business School. That study concludes, inter alia, that online adult entertainment subscriptions as a proportion of broadband households are: (1) more prevalent in U.S. states where religiosity is higher; and (2) weakly more prevalent in states where constitutional amendments banning same-sex marriage have been enacted.

In support of his study, Professor Edelman relies mainly upon three sources of data: (1) proprietary online adult entertainment subscription data from one of his consulting clients; (2) religiosity data from a Pew Research combined survey (1987-2007); and (3) voting data from 27 states where so-called "defense of marriage" amendments were enacted at the time of his research; the number of states enacting such amendments currently stands at 30. In an email exchange, Professor Edelman shared with me his subscription data for all 50 states and the District of Columbia as converted to nonproprietary client "subscriptions as a proportion of broadband households"--computations only partially cited in his study.

With the foregoing in mind, assuming arguendo that Professor Edelmen's subscription data is representative of online adult entertainment subscriptions and properly converted by him, his study fails to address an indispensable question; that is, is there a statistically significant correlation between U.S. states whose citizens are more religiously, socially and/or politically conservative--the so-called "Red States", and online adult entertainment subscriptions? If the correlation between those two variables is not statistically significant, the fact that Utah (the most prominent Mormon state) tops Professor Edelmen's porn per broadband list, is no more meaningful than Idaho (Utah's neighbor and the second most prominent Mormon state) coming in 50th place, or Tennessee (an extremely religious as well as socially and politically conservative state), coming in 49th place. Stated otherwise, at a minimum these two variables must be correlated in a statistically significant way before any conclusions can be drawn from the data.

For the reasons set forth below, there is no statistically significant correlation between U.S. states whose citizens are more religiously, socially and/or politically conservative and online adult entertainment subscriptions.

Null Hypothesis

There is no statistically significant correlation between U.S. states whose citizens are more religiously, socially and/or politically conservative and online adult entertainment subscriptions.

Statistical Correlation Measure

The statistical correlation measure used herein is the Pearson product-moment correlation coefficient (denoted by r) which measures the linear correlation between two variables X and Y (intervals), giving a value between -1 and +1. The closer the r statistic is to +1, the greater the positive correlation between two variables. The closer the r statistic is to -1, the greater the negative correlation between two variables.

For example, in the three studies below, an r statistic of +1 would indicate a perfect positive correlation between states with higher levels of religiosity, social and political conservatism and online adult entertainment subscriptions. By contrast, an r statistic of -1 would indicate a perfect negative correlation between states with higher levels of religiosity, social and political conservatism and online adult entertainment subscriptions. An r statistic of 0 indicates no statistical correlation.

Data

There are three correlation studies reported herein. In the first, X is an average measure of all eight religiosity criteria from the most recent Pew Research "U.S. Religious Landscape Survey" which was conducted from May to August 2007--a representative sample of more than 35,000 U.S. adults. Those eight religiosity criteria include: (1) belief in God absolutely; (2) considering religion to be very important; (3) attending church at least once a week; (4) praying at least once a day; (5) praying and receiving answers to prayers; (6) interpreting religious scriptures literally; (7) believing in only one interpretation of religious scriptures; and (8) believing a particular religion is the one true faith. For each of the eight criteria, percentages for the most dogmatic religious responses were used to calculate the average measure of religiosity for each state.

In the two remaining correlation studies, X is: (1) the percentage vote in favor of so-called "defense of marriage" amendments in each of the 30 states that have enacted those amendments; and (2) the percentage vote in each state for 2008 Republican presidential candidate, John McCain.

In all three studies, Y is Professor Edelman's online adult entertainment subscription data as a proportion of broadband households for each state and the District of Columbia.

Critical Value Range

(.27) to .27

Interpretation: In order to reject the null hypothesis above, the r statistic in each correlation study must be < (.27) or > .27. Any r statistic between (.27) and .27 indicates no statistically significant correlation between X and Y.

(N=51: 50 states plus D.C.; 49 degrees of freedom (51-2) at an alpha of .05).

Findings

(1) Religiosity and Online Adult Entertainment Subscriptions: r = .22
(2) Defense of Marriage Amendments (% vote) and Online Adult Entertainment Subscriptions: r = .20
(3) 2008 (% vote) for John McCain and Online Adult Entertainment Subscriptions: r = .19

Conclusion

The null hypothesis cannot be rejected because the r statistic for all three correlation studies is within the critical value range. There is no statistically significant correlation between U.S. states whose citizens are more religiously, socially and/or politically conservative and online adult entertainment subscriptions.

To read more from this author CLICK HERE.

Source(s):

"Critical Values of the Pearson Product-Moment Correlation Coefficient," University of Connecticut
Edelman, Benjamin, "Red Light States: Who Buys Adult Entertainment?" Journal of Economic Perspectives, Volume 23, Number 1 (Winter 2009)
Election Center-Presidential Race, CBS News
"U.S. Religious Landscape Survey," The Pew Forum on Religion and Public Life