Clergy Wives and Well-Being: The Impact of Perceived Congregational Perfectionism and Protective Factors

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is an important and much needed piece of research despite the convenience sample. I would support its publication after some minor revision along the following lines:

The major need is for clarity regarding the PPGS and its subscales. The instrument is not described in detail until the end of the ms, and it is only briefly described at the top of p. 3. The authors write, however, as if the names of and distinctions between the subscales of Standards and Discrepancy were already clear, even in abstract. It would be better even in the abstract to be clear terminologically that "perceived congregational perfectionism" has two correlated dimensions, and then to be as consistent as possible linguistically throughout the ms, even to the point of repetition. At the top of p. 3, "Perceived Standards from God" is italicized, but "perceived discrepancy" is not, making it even more confusing as to the distinction between them and whether the latter is a subscale. This is the place where it could be made clear. It would be particularly helpful, I think, to frame this in terms of perceived "judgment" from the congregation, since "discrepancy" unhelpfully suggests the possibility of a discrepancy score between two ratings...

Lines 5 and 38: "Hileman" is misspelled

Line 42: "wives" needs an apostrophe to be possessive

Lines 74-75: I would wish for the citation of Matt 5:48 to note that any reading of Jesus' teaching here that is used to support perfectionism is actually a misreading of the text and the overall tenor of the chapter

Lines 77-82: if possible, strengthen the connection between Hewitt/Flett and the literature on clergy wives

Line 94: I assume that by "perceived perfectionism" you mean "socially prescribed perfectionism," but you don't say this until later, and it would be helpful to make that clear here

Line 121: Similar to the earlier comment, I would wish for the reference to "biblical standards" and its link to an "ideal image" to be properly qualified to note that this is often a social misreading and misappropriation of the text

Lines 168-169: Here it might be helpful to cite the following: Lee, C. & Rosales, A. (2020), Self-regard in pastoral ministry: Self-compassion versus self-criticism in a sample of United Methodist clergy, Journal of Psychology and Theology, 48, 18-33.

Line 188: H1 only posits an association with Discrepancy, presumably because of the findings described briefly at the top of p. 3. But if a reader doesn't remember that, they'll be left wondering why only Discrepancy and not Standards is hypothesized here

Line 191: H2 looks for moderating effects on the DV of "congregational perfectionism"--but that has two scales, and H1 only mentions one

Line 230: I think you want "any" instead of "one"

Lines 233-234: this is a perfectly reasonable finding, but one that contradicts previous research, as will be said later. But it might be helpful to point out here that this is unexpected before highlighting it later

Line 390: "convenience" instead of "convenient"?

Line 428: "in both the English and Chinese languages"

Line 446: "Congregational" for consistency

 

 

Comments on the Quality of English Language

There are only a few places where editing is needed, and these are listed in my comments

Author Response

1.    The major need is for clarity regarding the PPGS and its subscales. The instrument is not described in detail until the end of the ms, and it is only briefly described at the top of p. 3. The authors write, however, as if the names of and distinctions between the subscales of Standards and Discrepancy were already clear, even in abstract. It would be better even in the abstract to be clear terminologically that "perceived congregational perfectionism" has two correlated dimensions, and then to be as consistent as possible linguistically throughout the ms, even to the point of repetition. At the top of p. 3, "Perceived Standards from God" is italicized, but "perceived discrepancy" is not, making it even more confusing as to the distinction between them and whether the latter is a subscale. This is the place where it could be made clear. It would be particularly helpful, I think, to frame this in terms of perceived "judgment" from the congregation, since "discrepancy" unhelpfully suggests the possibility of a discrepancy score between two ratings…
➔    We added a note in the abstract indicating the two PPGS dimensions. We have also italicized the perceived discrepancy on page 3. We’ve replaced the term “discrepancy” to “judgment.” Thank you for the suggestion.

2.    Lines 5 and 38: "Hileman" is misspelled
➔    fixed
 
3.    Line 42: "wives" needs an apostrophe to be possessive
➔    fixed

4.    Lines 74-75: I would wish for the citation of Matt 5:48 to note that any reading of Jesus' teaching here that is used to support perfectionism is actually a misreading of the text and the overall tenor of the chapter
➔    We added a citation and a sentence to note the social impact of the reading of the text, “although the definition of perfection varied based on the particular culture, tradition, and context (Clarke, 2003)”

5.    Lines 77-82: if possible, strengthen the connection between Hewitt/Flett and the literature on clergy wives
➔    We have now strengthened the connection by revising a few statements, “This former form of perfectionism involves holding unrealistic standards for significant others, placing a high value on their perfection, and rigorously assessing their performance, which can extend to the congregation's expectations of clergy wives. The latter refers to perceived perfectionism imposed by others and serves as an apt construct for understanding the high expectations clergy wives perceive from their congregational communities.” 

6.    Line 94: I assume that by "perceived perfectionism" you mean "socially prescribed perfectionism," but you don't say this until later, and it would be helpful to make that clear here
➔    Thank you, and we’ve made the change to enhance clarity. 

7.    Line 121: Similar to the earlier comment, I would wish for the reference to "biblical standards" and its link to an "ideal image" to be properly qualified to note that this is often a social misreading and misappropriation of the text
➔    We have added citations and noted the social misreading of the text, “defined through the lens of a particular social imagination rather than the scripture’s original intent for believers to imitate Jesus’ all- embracing love and seek maturity (Eaton, 2020; Stott, 2016)”

8.    Lines 168-169: Here it might be helpful to cite the following: Lee, C. & Rosales, A. (2020), Self-regard in pastoral ministry: Self-compassion versus self-criticism in a sample of United Methodist clergy, Journal of Psychology and Theology, 48, 18-33.
➔    Thank you. We have added the citation. 

9.    Line 188: H1 only posits an association with Discrepancy, presumably because of the findings described briefly at the top of p. 3. But if a reader doesn't remember that, they'll be left wondering why only Discrepancy and not Standards is hypothesized here
➔    We’ve now added a rationale for why we only made hypotheses for the Judgment dimension due it’s maladaptive nature. 

10.    Line 191: H2 looks for moderating effects on the DV of "congregational perfectionism"--but that has two scales, and H1 only mentions one
➔    We add a note that the associations between perceived congregational standards and the study variable were exploratory

11.    Line 230: I think you want "any" instead of "one"
➔    fixed

12.    Lines 233-234: this is a perfectly reasonable finding, but one that contradicts previous research, as will be said later. But it might be helpful to point out here that this is unexpected before highlighting it later
➔    Thank you, we’ve made edits to highlight this finding.  We now state, “The results on perceived standards contradicted previous research, as they showed that …”

13.    Line 390: "convenience" instead of "convenient"?
➔    fixed

14.    Line 428: "in both the English and Chinese languages"
➔    fixed

15.    Line 446: "Congregational" for consistency
➔     changed congregation’s to congregational

Reviewer 2 Report

Comments and Suggestions for Authors

The paper addresses an exciting issue, reviews the literature well, and formulates fair and well-supported hypotheses. The structure of the text is logical and easy to follow. However, in terms of interpretation of the results, there are two questions that would benefit from a slightly more detailed explanation.

1)        I agree with the author(s) of the article: pastors' wives are indeed an understudied population. A population whose specific situation (as a subordinate participant in a two-person-job) may have important gender-related lessons for the position of women in churches in general. However, as the position and opportunities of women in different Protestant churches differ, it would be important to know which churches / denominations are investigated. The authors - quite rightly - emphasize the limitations of their research due to the targeted population and the sampling method, but some questions arise which are not addressed. First, it is important to justify the choice of the target group. If I understand correctly, the study population is pastors' wives who self-identify as Asian all over the word. It is not self-evidently important (though it may be) in terms of the perceived expectations that the respondents are Asian, so it would be necessary to justify why they are the target group. At one point it reads "due to the majority of participants being Asian (...), non-Asian participants (...) were removed from the dataset” (213-214). So it seems that this wasn't the intention, it just happened that way? This should be clarified.

May the expectations be related to how the denomination thinks about the role of women in the church, and consequently: whether it ordains women? The extent to which traditional gender expectations are prevalent in a given community is obviously an important factor regarding perceived expectations, so it would be good to know how mixed or homogeneous the sample was in terms of denominationality and, if mixed, whether the impact of this factor was examined.

The perceived expectations may be influenced by the degree of the urbanization of the  environment in which the pastor's family lives - in a smaller settlement they are more likely to be traditionally in the eyes of the congregation, in a larger city the pressure of expectations may be weaker in this sense. I understand you have collected data on place of residence, so it would be worth noting whether you have examined this relationship.

The country / geographical location may also be important in terms of perceived expectations: these pastors' wives live in different countries around the world, most of them in Taiwan: have you investigated whether there are differences between them and those in other countries? We read "that perceived congregational perfectionism and well-being variables did not differ based on church size, and whether or not they hold a job outside of church or have children." (253-254) But we do not learn, why did the author(s) look only at these variables? May they not differ based on country, settlement type, denomination?

 

2)        The statistical analysis is very interesting and well structured. Nevertheless, the second question which needs to be clarified a little more is the stressed use of the p-value. As far as I know, p-values measure whether the differences observed in the sample are significant, i.e. whether they can be found (with a certain probability) in the whole population. However, probability sampling is a prerequisite for their use, their use on non-representative samples is at least debated. Though they are often used on non-probability samples, their use should be explained.

"The results showed that PCPS-Discrepancy had significantly stronger correlations with depression (p = .006), …”(241-243):  Here the formulation is somewhat misleading, since p certainly does not say anything about the strength of the correlation.

"Table 2. Summary of Moderation Effects Indicated through P-values.": It is not clear why the p-values are highlighted in a separate table. I would suggest to use the highlighting to draw attention to the strength of the moderating effect, not to the p value.

Some other minor comments:

"The pressure to adhere to biblical standards and portray an ideal image often hinders clergy wives from sharing personal struggles with their congregations, particularly regarding marital, family, and spiritual matters" (120-122) - the reason for this (and thus for loneliness) may be not only the need to adhere to biblical standards and portray an ideal image, but also the simple fact that this would mean sharing confidential information about their own pastor with members of the congregation, which many find unacceptable.

"there is a lack of research focusing on self-compassion and perfectionism among Christian leaders, particularly clergy wives." (168-169) Does this mean that the clergy wife is a leader? In what sense?

It would also be interesting to know whether the sample includes pastors' wives who are themselves pastors – which is a very specific situation. We learn that 28.8% of the participants reported being employed full-time and 9.8% part-time by their church, so it would be interesting to know if some of them are pastors themselves.

All in all, I enjoyed reading the study. The points indicated do not need correction so much, rather they need further explanation, clarification or in some cases justification. But in my opinion, these are definitely needed.

Author Response

1.    1)        I agree with the author(s) of the article: pastors' wives are indeed an understudied population. A population whose specific situation (as a subordinate participant in a two-person-job) may have important gender-related lessons for the position of women in churches in general. However, as the position and opportunities of women in different Protestant churches differ, it would be important to know which churches / denominations are investigated. The authors - quite rightly - emphasize the limitations of their research due to the targeted population and the sampling method, but some questions arise which are not addressed. First, it is important to justify the choice of the target group. If I understand correctly, the study population is pastors' wives who self-identify as Asian all over the word. It is not self-evidently important (though it may be) in terms of the perceived expectations that the respondents are Asian, so it would be necessary to justify why they are the target group. At one point it reads "due to the majority of participants being Asian (...), non-Asian participants (...) were removed from the dataset” (213-214). So it seems that this wasn't the intention, it just happened that way? This should be clarified.
➔    We included this sentence to clarify the denominations represented: There were more than 30 denominations/traditions represented in this study, including Presbyteria, Baptise, Lutheran, Bread of Life, and Independent churches.”
➔    For data collection, we did not indicate race as an eligibility criterion. However, using a snowball approach, the final sample ended up heavily with Asian participants due to the researcher’s connection.  
➔    The limitations of the convenience sampling and predominantly Taiwanese clergy wives are noted in the limitation section. 

2.    May the expectations be related to how the denomination thinks about the role of women in the church, and consequently: whether it ordains women? The extent to which traditional gender expectations are prevalent in a given community is obviously an important factor regarding perceived expectations, so it would be good to know how mixed or homogeneous the sample was in terms of denominationality and, if mixed, whether the impact of this factor was examined.
➔    Thank you for raising this important question. Unfortunately, although we have basic information regarding the denominations represented, we did not ask about ordination. We do not want to make assumptions based on how Western churches operate, given that the majority of the participants were in Taiwan. However, we believe that this is worth further understanding in future studies and have added this consideration in the future directions. 

3.    The perceived expectations may be influenced by the degree of the urbanization of the environment in which the pastor's family lives - in a smaller settlement they are more likely to be traditionally in the eyes of the congregation, in a larger city the pressure of expectations may be weaker in this sense. I understand you have collected data on place of residence, so it would be worth noting whether you have examined this relationship.
➔    Thank you for this point. We have considered examining this relationship. However, given that the majority of the participants were from Taiwan, where the definition of “city” is very different from that of the West, and considering that the definitions of a “larger” city and a “small” one can be arbitrary, we believe the results would be complicated and potentially misleading based on the cities reported by the participants. In other words, with the data that we collected based on our study design, this valid question would be difficult to adequately answer.

4.    The country / geographical location may also be important in terms of perceived expectations: these pastors' wives live in different countries around the world, most of them in Taiwan: have you investigated whether there are differences between them and those in other countries? We read "that perceived congregational perfectionism and well-being variables did not differ based on church size, and whether or not they hold a job outside of church or have children." (253-254) But we do not learn, why did the author(s) look only at these variables? May they not differ based on country, settlement type, denomination?
➔    This is a good question.  However, a key challenge is the varied sample sizes in country groups and lumping could potentially create heterogeneity concerns. As for the denomination data, it was collected through an open question (rather than predefined categories), so it would be challenging to adequately categorize and investigate the question of denominational difference. 
Similarly, the sample size variation is also an issue commonly found in examining denominational differences. So, these questions were not quantitatively analyzed in this study.
 
5.    2)        The statistical analysis is very interesting and well structured. Nevertheless, the second question which needs to be clarified a little more is the stressed use of the p-value. As far as I know, p-values measure whether the differences observed in the sample are significant, i.e. whether they can be found (with a certain probability) in the whole population. However, probability sampling is a prerequisite for their use, their use on non-representative samples is at least debated. Though they are often used on non-probability samples, their use should be explained.
➔    The p-values for the correlations and moderation effects reflect what was found among the variables examined in this sample. More specifically, the significant p-values reflect statistical relationships that are different from zero (absolutely no correlation). Since the findings pertain to this sample, generalizability is limited, which we now note in the limitation section. “This could limit generalizability to the larger clergy-wives population from both ethnicity and non-representative sampling perspectives.”

6.    "The results showed that PCPS-Discrepancy had significantly stronger correlations with depression (p = .006), …”(241-243):  Here the formulation is somewhat misleading, since p certainly does not say anything about the strength of the correlation.
➔    The comparison of strength for each pair of corresponding correlations (Standard vs Judgment) yielded Z-scores that were compared in a 2-tailed fashion to the unit normal distribution.  P-values indicate whether there were statistically significant differences in how the two PCPS dimensions correlated with a certain study variable.
➔    We have revised the manuscript to enhance clarity. We now state, “Lastly, an asymptotic z-test of the difference between the relative strength of correlations of the two PCPS subscales (i.e., PCPS-Standards and PCPS-Judgment) with study variables was conducted using Lee & Preacher’s (2013) online calculator. Z-scores were compared in a 2-tailed fashion to the unit normal distribution, and p-values indicate whether there were statistically significant differences in how the two PCPS dimensions correlated with a certain study variable.”

7.    "Table 2. Summary of Moderation Effects Indicated through P-values.": It is not clear why the p-values are highlighted in a separate table. I would suggest to use the highlighting to draw attention to the strength of the moderating effect, not to the p value.
➔    Based on our understanding, the p-values reflect the significance of the moderation effect. As we do note the B and SE in the text, we’ve now removed the p-values table.

Some other minor comments:
8.    "The pressure to adhere to biblical standards and portray an ideal image often hinders clergy wives from sharing personal struggles with their congregations, particularly regarding marital, family, and spiritual matters" (120-122) - the reason for this (and thus for loneliness) may be not only the need to adhere to biblical standards and portray an ideal image, but also the simple fact that this would mean sharing confidential information about their own pastor with members of the congregation, which many find unacceptable.
➔    That is a great point. We’ve included the confidentiality consideration mentioned in the literature. We state, “Luedtke and Sneed (2018) found that, in addition to perceived expectations, most participants described their experience in ministry as “lonely,” as there is also a perceived need to be extra cautious with confidential information.

9.    "there is a lack of research focusing on self-compassion and perfectionism among Christian leaders, particularly clergy wives." (168-169) Does this mean that the clergy wife is a leader? In what sense?
➔    We have added another citation and made edits to remove any confusion and specify the need for research specifically on clergy wives. 

10.    It would also be interesting to know whether the sample includes pastors' wives who are themselves pastors – which is a very specific situation. We learn that 28.8% of the participants reported being employed full-time and 9.8% part-time by their church, so it would be interesting to know if some of them are pastors themselves.
➔    Thank you for raising this question. Unfortunately, we did not ask participants to specify if they were pastors themselves, so we do not have this information. However, we have made edits in the future directions section to address this consideration.