Adventist Media Response and Conversation

Saturday, September 28, 2013

Lessons in the art of survey manipulation

I am going to try a new tack on this blog I often read comments and articles that I would like to comment on but really practically no one reads the comments sections so I thought I would use them as short blog articles.

So here is the first one from Spectrum a comment writer says:
"conservative Christians are the only ones who read the bible seriously"
That's funny because the only data based study I've ever seen on the subject suggests that liberals are far more likely to read their bible. The thing is many regressive Christians are very good at memorizing a verse here and a verse their to support their pre-established suppositions, where as most progressives actually read and understand."
Here is a section from the likely source of the above thinking. An article from Christianity Today, Survey: Frequent Bible Reading Can Turn You LiberalWhat a surprising survey says about how reading the Bible frequently can turn you liberal (in some ways):  
 Frequent Bible reading has some predictable effects on the reader. It increases opposition to abortion     as    well as homosexual marriage and unions. It boosts a belief that science helps reveal God's glory. It diminishes hopes that science will eventually solve humanity's problems. But unlike some other religious practices, reading the Bible more often has some liberalizing effects—or at least makes the reader more prone to agree with liberals on certain issues. This is true even when accounting for factors such as political beliefs, education level, income level, gender, race, and religious measures (like which religious tradition one affiliates with, and one's views of biblical literalism).
In 2007, the Baylor Religion Survey asked Americans how often they read the Bible on their own. (It was a five-point scale in this study, ranging from "never" to "several times a week.") It also asked whether the federal government should expand its authority to fight terrorism—a reference to the Patriot Act. For each increased level of Bible-reading frequency, support for the Patriot Act decreased by about 13 percent.

Frequent Bible reading also influences views on criminal justice. As might be expected, respondents who were more politically liberal were prone to disagree with the statement, "The government should punish criminals more harshly." Unexpectedly (at least given the conservative stereotype), the more frequently people read the Bible, the more they too are prone to disagree with the statement. This is not an anomalous finding: Support for abolishing the death penalty increased by about 45 percent for each increase on the five-point scale measuring Bible-reading frequency.

Reading the Bible affects attitudes toward science as well. If you just ask people about biblical literalism, you don't find statistically significant differences in views of whether science and religion are compatible. But the more someone reads the Bible, the more likely he or she is to believe science and religion are compatible. (For each increase on the five-point scale, the odds that they see religion and science as incompatible decrease by 22 percent.)
Some of the most interesting findings relate to moral attitudes. "How important is it," the survey asked, "to actively seek social and economic justice in order to be a good person?" Again, as would be expected, those with more liberal political leanings were more likely to say it's very or somewhat important. And those who read the Bible more often were more likely to agree. Indeed, they were almost 35 percent more likely to agree at each point on Baylor's five-point scale. That may be bad news for Glenn Beck, who last year told believers to leave their churches if they hear "social justice" language being used. Likewise, contrary to liberal media stereotypes, those who are most engaged in their faith (by directly and frequently reading its source material) are those who are most supportive of social and economic justice. A reading, politically conservative literalist is only slightly less supportive than a non-reading, politically liberal non-literalist.

Now just so you don't get the idea that there is not a liberal agenda in Christianity Today consider the last part about Glenn Beck. Beck was specific about what "Social Justice" he was referring too. See: Glenn Beck: What is Social Justice?

"What is that? It seems like such an innocuous phrase. It paints a picture of fairness — many churches use the term as a substitute for "outreach to the poor." Who could possibly be against that? Well, if you’ve read the news lately: I am. In fact, I even learned from TIME magazine recently that I hate Jesus.

I’m just full of hate and I want to stop justice!

I’m glad to see Time suddenly cares about God… or am I? The other "news" from The New York Times was that I recommended leaving church if those churches help the poor. And I would have gotten away with it too, if it wasn’t for those pesky, meddling "journalists"!

I’m not sure why I would expect the media to start searching for the truth now, when they’ve never let it get in the way before. The truth is this: The term "social justice" has been completely perverted and hijacked by progressives. It doesn’t mean simply "help the poor" to them. It does to some people, but not to radical progressives.

And now, just for The New York Times and everyone else who thinks I hate poor people — I know your attention span is about 20 or 30 seconds, but try and pay attention — we’ll set the record straight for you here on social justice. The kind I am talking about vs. the kind that they are talking about.


Here’s my definition of social justice: Forced redistribution of wealth with a hostility toward individual property rights, under the guise of charity and/or justice.

On my radio program, I said if your church is promoting Jeremiah Wright-type "social or economic justice," you should run from it or at least get educated on what progressives mean by this.

You might wonder if the summary from Christianity Today's writer is applicable. In particular since the summary of the poll does not mention the correlations that the writer does namely: In 2007, the Baylor Religion Survey asked Americans how often they read the Bible on their own. (It was a five-point scale in this study, ranging from "never" to "several times a week.") It also asked whether the federal government should expand its authority to fight terrorism—a reference to the Patriot Act. For each increased level of Bible-reading frequency, support for the Patriot Act decreased by about 13 percent.

The survey codebook is found here.

56) Q16. Outside of attending religious services, about how often do you read the Bible, Koran, Torah, or other sacred book? (SACREDBK)

0) Never44427.3
1) Less than once a year23214.3
2) Once or twice a year18011.1
3) Several times a year16410.1
4) Once a month493
5) 2-3 times a month986
6) About weekly925.7
7) Weekly1157.1
8) Several times a week or more often25415.6

172) Q37h. To what extent do you agree or disagree that the federal government should expand its authority to fight terrorism? (FIGHTTER)

1) Strongly disagree1418.8
2) Disagree38824.3
3) Agree55534.7
4) Strongly agree39524.7
8) Undecided1217.6

If you look at the analyze section of question 172 You actually see that there is no statistics related to Bible reading. It correlates Age, Education, Gender, Religion, Region and Church attendance.

It is interesting to see how one article with really questionable correlations can affect those that want to believe something.


Anonymous said...

Not to mention the more basic flaw of the CT author who mistakes a correlation for cause and effect relationship, thereby concluding that more frequent Bible reading "causes" liberalism! i.e., Bible reading has predictable effects on the reader, it increases opposition to abortion, etc. No chance it is the other way around? People opposed to abortion are more likely to be frequent Bible readers? The CT article is full of such sloppy analysis/language.

Ron Corson said...

I looked at the comments on the CT article and it does not seem to be that most people noticed the artificial correlations that the writer drew. Most probably did not even see the poll, after all Frazen did not even post a link to it and never posted one statistic, just his claim of percentage changes. In any case the first letter I think is a pretty good response if one assumes there is actually some facts behind the article:

" Morgan
October 12, 2011

I think this article is misleading and skewed in one sense. The author acknowledged that people who read the Bible are more likely to oppose abortion and homosexual marriage. So on those issues reading the Bible apparently makes one more conservative. However, this aspect wasn't acknowledged in the article, making it seem as if the author had a liberalizing agenda. Based on the author's own statements, a more accurate claim would be that the Bible makes people more liberal on certain topics but more conservative on others."