Get a FREE copy of The Secular Creed by Rebecca McLaughlin


The Story: A new study says Calvinists are prone to believe “myths” that may lead them to justify domestic violence. But is there evidence to support this claim?

The Background:  Psychologist Steven Sandage, a professor of psychology of religion and theology at Boston University, recently published a study in the Journal of Psychology and Spirituality, which implies that Calvinism sometimes justifies or rationalizes violence against women.

In an interview with BU Today, Sandage summarizes his research by saying, “Many Christian theologies emphasize the possibility of finding meaning in suffering, but the New Calvinism seems to promote a rather stoic and un-empathic attitude that valorizes suffering, particularly among women. . . Calvinist beliefs were related to higher levels of domestic violence myth acceptance and lower levels of social justice commitment.”

In the Calvinist view, “God causes all things, including hierarchical social structures and all suffering,” Sandage says. “Domination by the powerful,” be it God or men, “is just and appropriate, and submission to suffering by the less powerful is virtuous and redemptive.”

Sandage is quick to add that not all Calvinists endorse domestic violence myths: “There are many contemporary Calvinists who hold progressive views of gender and other social issues. But our research does offer some data suggesting the ‘New Calvinism’ that combines Calvinistic beliefs and very conservative, binary views of gender may be a kind of theological risk factor for the acceptance of domestic violence myths and other socially regressive attitudes.”

The BU Today article also notes, “Scot McKnight, a professor of New Testament at Northern Seminary in Illinois, lauds Sandage’s research for drawing ‘accurate and helpful correlations that ought to awaken more theologians and pastors to the implications’ of their theology.

What It Means: [UPDATE: After this article was published I had a friendly and productive exchange with Dr. Sandage. The sections in brackets include updates and corrections based on my discussion with him.]

When I first heard about (and read) the study, I didn’t give it much thought. It’s the type of social science research that uses biased assumptions and flawed “tools” to generate unsupportable conclusions that can clear the low threshold necessary for publication in a middle-tier academic journal. In the publish-or-perish world of academia, this type of low-quality, never-to-be-replicated work is often the norm. [Update: Dr. Sandage took issue my claim that the study is “never-to-be-replicated.” As he notes, this is “obviously a future-oriented statement.” That’s certainly true, and I apologize for any misperception. If any subsequent studies replicate the findings I will include them in a future update.]

But then I saw that some well-meaning people, such as the respected scholar Scot McKnight, were citing the study favorably, and even as important finding. I’m not sure if those, like McKnight, who agree with the study are also embracing radical feminist theory, whether they didn’t read the study all that closely, or whether they simply don’t understand the theory behind the claims of the study. For whatever reason, I figured that if reasonable Christians were falling for the spurious conclusions of the study then it might be worth addressing in detail.

As you might imagine, someone who think Calvinists believe “domination by the powerful . . . is just and appropriate” is not likely to be a reliable guide to Calvinistic beliefs. Initially, though, I assumed Sandage and his team were simply afflicted with a virulent case of anti-Calvinism. But it turns out that it’s not so much theological beliefs as an embrace of radical feminist theory that is the driving impetus of the study.

The study itself is rather shoddy and could be picked apart from many angles. But the simplest way to show why is it unreliable is to point out the flaws in its primary diagnostic tool, the Domestic Violence Myth Acceptance Scale.

The Domestic Violence Myth Acceptance Scale (DVMAS) was created by John Peters in 2003 as part of his doctoral thesis in social work. It uses 18 statements to gauge whether a person believes “myths” about domestic violence. To show why this study is unreliable, I’ll be pointing out five main flaws that undermine the credibility of the scale.

First, Peters defines a “myth” as “stereotypical attitudes and beliefs that are generally false.” But that is not the definition of a myth, which is always false, not “generally false.” If there are occasions when a claim is sometimes true, then we must understand the context to determine whether the assertion is true or false. But the DVMAS relies on ambiguous claims about “myths” that exclude any nuance or context and thereby negates the usefulness of it as a diagnostic tool.

Second, the basis for the DVMAS is radical feminist theory. As Peters says in his thesis,

The radical feminist model . . . contends that the violence supports and is supported by patriarchal oppression of women. This model of violence resulting from patriarchal socialization implies that rape, domestic violence, and other forms of violence against woman are part of broader social attitudes toward women.

As we will see later in this article, there is a paucity of empirical support for the radical feminist model of domestic violence.

Third, Peters defines domestic violence as, “Violence between intimate partners which has as its goal establishing and maintaining a culturally sanctioned pattern of power and control by men over women within the context of an intimate relationship.” While that definition is fitting under radical feminist theory, most Americans have a broader view of domestic violence. This difference in how the primary concept is understood is likely to skew the results.

Fourth, the scale is intended to be a “reliable and valid measure of [DV] myths.” But Peters has no way of determining whether the answers to the 18 statements are in any way connected to actual beliefs about domestic violence. Instead, he simply measures how they are correlated with several other scales, including the Burt’s Sex-Role Stereotype scale [1980] (“a well-validated measure of sexual conservatism which has been shown to be highly correlated with rape myths and with negative attitudes toward domestic violence victims”) and the Attitudes Toward Woman Scale [1974] (“a unifactorial measure of both sex-role conservatism and general attitudes toward women (Spence, Helmreich, & Stapp, 1974). Rather than being a reliable measure of what people today believe about domestic violence, the scale merely reflects possible correlations between the DMVAS and other (similarly flawed) scales intended to measure “sexually conservative” views held in the 1970s.

[Update: Dr. Sandage says that none of the items from the original 18-item DVMAS discussed below were items used in their study. Instead, they used an 8-item version based on “Jankowski’s factor analytic work.” The eight questions they used were:

If a woman continues living with a man who beat her, then its her own fault if she is beaten again

Making a man jealous is asking for it.

A lot of domestic violence occurs because women keep on arguing about things with their partners.

Women who flirt are asking for it.

Women can avoid physical abuse if they give in occasionally.

Many women have an unconscious wish to be dominated by their partners.

Domestic violence results from a momentary loss of temper.

Women instigate most family violence

While I believe these questions suffer from the same weaknesses and flaws as the ones I critique below, I believe it is important to be as accurate as possible. I appreciate Dr. Sandage providing this information.]

Fifth, the DVMAS asks respondents whether they agree or disagree with common statements about domestic violence. Each of these 18 statements are supposedly representative of a “myth,” as defined by Peters:

For this study, domestic violence myths were therefore defined as statements about domestic violence which invoke either character blame of the victim, behavioral blame of the victim, exoneration of the perpetrator, or minimization of the seriousness or extent of the problem.

Nowhere in his thesis, however, does Peters explain how he determined that these statements, using the wording he crafted, are proven to be reliable guides to ascertaining whether a person believes a “domestic violence myth.” He also doesn’t directly back up the statements with any empirical research. He simply expects that anyone who starts from the “correct” perspective (i.e., a radical feminist view) will answer in the way he wants them to.

Because this is the fatal flaw that destroys the credibility and usefulness of the scale, let’s examine a few of the statements.

Statement #8 on the DVMAS is that, “Most domestic violence involves mutual violence between the partners.”

This is a prime example of the way the scale combines ambiguous language (e.g., “most”) and empirically questionable assumptions to devise a question where political correctness trumps more rigorous research on domestic violence. It’s also an attempt by those who use the scale to dismiss anyone familiar with the actual empirical data on such issues as gender symmetry in the perpetuation of domestic violence.

For example, in 2010, Murray A. Straus wrote an article for the journal Partner Abuse in which he says,

Although at least 200 papers report research that found gender symmetry in perpetration, many studies with similar results were not submitted for publication because the authors thought a paper showing gender symmetry would not be accepted or because the authors feared adverse effects on their reputation and employability.

(Note: As Straus also notes in the paper, the reality of gender symmetry doesn’t negate the reality that there is not a corresponding symmetry in power. While domestic violence is harmful no matter who initiates the conflict, the physical power imbalance means that the affect of such violence on women is usually more severe. Recognizing that both genders may initiate conflict in no way excuses the behavior of men who engage in violence against women.)

In referring to early studies that had been ignored, Straus says,

Why were these statistics presented and the implications ignored? An important part of the explanation was that these results contradicted the feminist analysis of [partner abuse] that had made both the academic world and the general public conscious of [partner abuse].

This same criticism applies to the DVMAS: it delegitimizes any perspective on domestic violence that disagrees with or contradicts radical feminist theory. This is evident in several other statements, such as, “When a man is violent it is because he lost control of his temper.”

For the DVMAS, this is considered a “myth” even though the claim is clearly supported by empirical research. As Erica Birkleya and Christopher I. Eckhardta wrote in a 2015 meta-analytic review of current research on intimate partner violence (IVP):

There currently exists a substantial, and hotly contested, debate in the [intimate partner violence] field about whether anger has any meaningful relation to IVP, whether anger-related constructs should be included in assessment for [intimate partner violence] risk, and whether anger-related variables should be the focus of IVP interventions to any degree. Much of this debate stems from assumptions based on the earliest, and still currently popular, model of IPV etiology: Power and control theory. This model, which is the predominant perspective in the broader IPV field, focuses exclusively on gender socialization patterns and defines IPV as male-to-female violence deeply rooted in gender-based power dynamics that play out in the romantic context. . . . Thus, adherents to this model place little emphasis on factors internal to the individual (such as anger or other negative emotions) as causes of behavior, preferring instead an analysis of community and contextual-based determinants of power-and-control socialization patterns.

The DVMAS is based on just this sort of “power and control theory,” and considers anyone who believes factors internal to an individual (such as anger) are a factor in domestic violence to be embracing a “myth.” The DVMAS thereby demonizes anyone who does not think that “gender-based power dynamics” are the sole factors that should be considered in understanding and preventing domestic violence.

This radical feminist view is especially problematic since it has no empirical support. As Birkleya and Eckhardta point out,

[T]here is very little empirical support for a strictly gendered analysis of IPV that restricts the understanding of IPV to the behaviors enacted by men towards women, or that organizes IPV risk factors solely around gender-themed attitudes or behaviors, especially as proximal causes of IPV-related outcomes. Rather, the available data suggest a gender-inclusive approach to IPV etiology that considers a wide range of individual, interpersonal, and contextual risk factors that may lead both men and women to act aggressively towards an intimate. Of relevance to this report, several theoretical models appear to offer support for anger, hostility, and internalizing negative affect as important risk factors for IPV perpetration. [emphases in original]

Out of the 18 statements on the DVMAS, three of them (18 percent) are about anger. They are all claims that a reasonable person could believe to be true and yet would not consider to be in any way a justification or downplaying of domestic violence. (The other two questions are, “Abusive men lose control so much that they don’t know what they’re doing” and, “Domestic violence results from a momentary loss of temper.” Again, both of the statement could be empirically true and yet have no bearing on what a person believes about domestic violence.)

Some of the statements are similarly biased. For example, statement #7 is, “If a woman doesn’t like it, she can leave.” This is a strangely worded question that reveals the general incoherence and ambiguity of the DVMAS. What is the proper answer? The respondent is expected to reply that a way is unable to leave, a response that denies the agency of all women. It is certainly true that there are many situations in which an abused women, for a wide variety of reasons, feels she is unable to leave her abuser. But women do leave—it is not a “myth.” The idea that women in general are unable to leave is not only empirically false but also deeply misogynistic.

Two other statements on the DVMAS are based on ambiguous terms that require making assumptions based on estimations: “Domestic violence does not affect many people” and “Domestic violence rarely happens in my neighborhood.”

In the first statement, the respondent is expected to determine what is considered “many people.” While they might be wrong, it’s possible that someone could think that domestic violence is a serious problem that nevertheless does not affect “many” people (based on their own conception of what constitutes “many”). In the second statement, the respondent is expected to guess what is meant by “rarely” and “neighborhood.” Domestic violence is based on socioeconomic factors that are not evenly distributed in all areas of the country, so it is numerically possible (and even quite likely) that there will be areas that could be construed as “neighborhoods” in which domestic violence “rarely happens.” There’s no reason to assume that giving the “wrong” answer on these questions is a sign that a person is believing a “myth” about domestic violence that would cause them to downplay or dismiss the problem. Even if we have a strong opinion about how people ought to answer these two questions, we should be able to agree that the questions are not a reliable measure of beliefs.

Out of the 18 statements, at least four are contradicted by empirical evidence, one is a bizarre claim that denies reality, and two are hopelessly ambiguous. There is no justification—at least not for anyone who doesn’t have an idealogical ax to grind—for trusting a measurement where between 22 percent and 39 percent of the scale is flawed.


What the paper by Sandage, et al., shows is that there is (potentially) a negative correlation between Calvinism (which the study may or may not accurately measure) and uncritical acceptance of the radical feminist view of domestic violence. It tells us nothing about how Calvinists actually view domestic violence myths, much less their view on the acceptability of domestic violence. Yet this work will lead some people to believe it has found a connection, and lead other people to assume that all modern psychological research is as biased and worthy of dismissal as this study.

What makes it even worse is this sloppy and biased study was produced by scholars who are Christians; it might have been more reliable had it been produced by Christian scholars. The distinction, as Nicholas Wolterstorff once said, is that, “To put it in a nutshell, I think the project of being a Christian scholar is the project of thinking with a Christian mind and speaking with a Christian voice within your chosen discipline and within the academy more generally.”

There is nothing in this study that bears evidence of having come from a “Christian mind,” nor does it speak with a “Christian voice.” In fact, there is nothing in the research that could not have been produced by any non-Christian radical feminist scholar. For that reason, and all the others cited in this article, there is no grounds for Christians—or anyone else who doesn’t embrace the intellectual fad of radical feminism—to take the study seriously, or to consider the findings anything other than a smear against Reformed believers.