Abstract
Romantic perfectionismi can be disruptive to relationships, yet no validated measure for assessing romantic perfectionism in Iranian couples has been developed. Therefore, the purpose of this study was to translate and validate the Romantic Perfectionism Scale (RPS) among Iranian couples. Participants in the study were 200 married men and 320 married women from Tehran, Iran, who completed the translated RPS, the Almost Perfect Scale-Revised, and the Depression Anxiety Stress Scale-21 online. Item impact scores were used to calculate face validity. Impact score values for all items were greater than 1.5, signaling appropriate face validity.. The Content Validity Index (CVI) and the Content Validity Ratio (CVR) were used to measure content validity. Values of the CVI were above the cut-off score of 0.7, implying satisfactory content validity of the items. The CVR values were greater than the Lawshe table (0.78) cut-off score, demonstrating that all items were essential. Confirmatory Factor Analysis (CFA) using AMOS software was used to evaluate the construct validity. The results of the goodness of fit indices confirmed the RPS with two subscales (i.e., self-oriented romantic perfectionism and other-oriented romantic perfectionism) as per the original scale. All items remained in the scale as all factor loading values were greater than 0.45. The findings showed that the two subscales, and the scale as a whole, had acceptable internal consistency, as the construct reliability values for self-oriented romantic perfectionism (0.81), other-oriented romantic perfectionism (0.72), and the whole scale (0.74) were greater than 0.7. The results support the psychometric properties of the Iranian version of the RPS, which could be used by future researchers and clinicians to assess romantic perfectionism in Iranian couples.
Introduction
The number of studies investigating perfectionism has grown in recent times and this may be due toincreasingly challenging and competitive environments as well as perceptions of higher standards and demands imposed on young people compared to generations past (Curran & Hill, 2019). To date, studies have found associations between perfectionism and a number of serious intrapersonal and interpersonal problems. Perfectionism has been found to be associated with depression (Abdollahi, Allen, et al., 2020), suicidal ideation (Abdollahi & Carlbring, 2017), rumination, social anxiety (Abdollahi, 2019), academic procrastination (Abdollahi, Farab, et al., 2020), test anxiety (Abdollahi et al., 2018), narcissistic personality disorder (Casale et al., 2020), and borderline personality disorder (Chen et al., 2019).
Perfectionism has negative implications for romantic relationships (Stoeber, 2012), and is associated with reduced sexual satisfaction, fear of intimacy (Habke et al., 1999; Martin & Ashby, 2004), conflict between couples, less forgiveness between couples (Furman et al., 2017; Mackinnon et al., 2012), and negative marital satisfaction (Kim et al., 2011). One study found that perfectionistic self-presentation was associated with low dyadic adjustment as well as self-conscious anxiety and obsessive preoccupations in romantic relationships (Flett et al., 2001).
A number of theories have attempted to describe the purpose of perfectionism. For example, Hamachek (1978) hypothesized that individuals with perfectionistic tendencies can be classified in two groups. 1. Adaptive (normal) perfectionists who have high but somewhat flexible standards, and 2. Maladaptive (neurotic) perfectionists who have extremely high and inflexible standards. Perfectionism has also been conceptualized as being positive or negative (Terry-Short et al., 1995). Positive perfectionism is correlated with an individual’s need to seek positive encouragement, whereas negative perfectionism is linked to an individual’s urge to prevent aversive effects (Terry-Short et al., 1995). Further, Hewitt and Flett (2002) posited that the difference between adaptive and maladaptive perfectionism depends on the degree of perfectionism (e.g., high or low), rather than on the form of perfectionism (e.g., positive or negative).
More recently, researchers have taken a multi-dimensional view of perfectionism that contains both adaptive and maladaptive characteristics, however there is disagreement on how perfectionism is operationalized. For instance, Frost et al. (1990) developed the Frost Multidimensional Perfectionism Scale and identified six dimensions of perfectionism: high standard expectations, doubts about behavior, concerns about failures, organization, parental expectations, and parental criticism. Hewitt and Flett (1989), on the other hand, identified three dimensions of perfectionism, including self-oriented perfectionism, other-oriented perfectionism, and socially prescribed perfectionism.
Previous studies have indicated that perfectionists have a tendency to respond defensively to criticism, cover imperfections, and apply unrealistic expectations on others. These features of perfectionism can be destructive for interpersonal relationships and lead to relationship dissatisfaction, particularly for married couples (Kim et al., 2011; Mee et al., 2015). Studies have also shown that other-oriented perfectionism (i.e., having high expectations for others) is associated with low levels of sexual satisfaction and exploitative and oppressive behaviors with a tendency to blame one's spouse (Kim et al., 2011; Mee et al., 2015). Self-oriented perfectionism is also associated with criticism, sarcasm, and retaliation in interpersonal relationships that negatively impact relationships (Stoeber, 2014).
While research has focused on the consequences of perfectionism within romantic relationships, the majority of studies have utilized scales intended to capture general perfectionism, rather than romantic-related perfectionism (e.g., Kim et al., 2011). To date, only one measure, the Romantic Perfectionism Scale (RPS; Matte & Lafontaine, 2012) has been developed, validated, and published to specifically assess romantic perfectionism. This measure is based on the 3-factor conceptualization of perfectionism by Hewitt and Flett (1989), however the scale only focuses on self- and other-oriented romantic perfectionism (Matte & Lafontaine, 2012). The advantage of the RPS over other scales (e.g., the Frost Multidimensional Perfectionism Scale) is that the scale focuses directly on perfectionism within romantic relationships rather than on perfectionism generally. Given the critical impact of romantic perfectionism in relationships (Furman et al., 2017; Habke et al., 1999; Mackinnon et al., 2012; Martin & Ashby, 2004), validation of the RPS for the Iranian population is necessary with therapeutic implications for clinicians who work with couples or practice family therapy. As the RPS has yet to be translated and validated in the Persian language, the purpose of this research was to translate and examine the psychometric properties of the RPS in a sample of married Iranian participants.
Method
Participants
A total of 537 participants completed the survey. After removing 17 respondents who were not eligible for the study (i.e., married less than one year), a total of 520 respondents were included, of which 320 (61.5%) were women and 200 (38.5%) were men. Participants ranged in age from 18 to 50 years (M = 29.12, SD = 6.45), and had been married on average for 4 years (SD = 2.4). All participants were Iranian, lived in Tehran, and cohabitated. Regarding education, 40 participants (7.7%) did not have a diploma, 120 respondents (23.1%) had a High School Diploma, 280 respondents (53.8%) had a Bachelor's degree, 70 respondents (13.5%) had a Master's degree, and 10 respondents (1.9%) held a Doctorate degree.
Procedure
The ethics committee of Alzahra University accepted the procedure and research materials of the present study. Questionnaires were developed and presented in a Google Form and an online link was shared via WhatsApp and Telegram platforms to potential respondents. All respondents of the online survey were living in Tehran, Iran. After agreeing to participate in the study by accepting an informed consent form, participants were informed that their information would remain confidential and were given the option to receive the research results after the study was concluded. Inclusion criteria to participate included the following: participants must be married for at least 12 months, be in a heterosexual relationship, and live in Tehran, Iran. Data collection took place in December 2020 and participants took an average of 15 minutes to complete the survey. No incentive was given to respondents for completing the questionnaire.
Instruments
Romantic Perfectionism Scale (RPS; Matte & Lafontaine, 2012). This measure consists of 14 items and evaluates two dimensions of romantic perfectionism, including self- and other-oriented romantic perfectionism. The self-oriented sub-scale evaluates proper behavior (3 items) and mistakes (4 items). Other-oriented romantic perfectionism measures disappointment of partner’s behavior (4 items) and expectations towards one’s partner (3 items). The measure utilizes a 7-point Likert scale from 1 (very strongly disagree) to 7 (very strongly agree), with a greater score indicating a greater level of the perfectionism. A previous study reported an acceptable internal consistency with a Cronbach’s Alpha of 0.76 (Medvedev et al., 2020).
The Brislin (1986) translation method was employed in order to translate the English version of the RPS into Persian, which utilized two independent translators. The translation of the RPS measure from English into Persian was completed by one translator. The second translator then independently back-translated the Persian RPS measure from Persian into English. Finally, the two versions were compared by three independent translators and no major variations in terms of content and concept between the Persian version and the original version were found.
Almost Perfect Scale-Revised (Slaney et al., 2001). This measure consists of 23 items and evaluates individual perfectionism with three subscales: discrepancy, order, and standards. The discrepancy subscale is evaluated with 12 items (e.g., Doing my best never seems to be enough), the order subscale is evaluated with 4 items (e.g., I am an orderly person), and the standards subscale is evaluated with 7 items (e.g., I have high expectations for myself). This measure utilizes a 7-point Likert scale from 1 (strongly disagree) to 7 (strongly agree), with a higher score indicating a greater level of perfectionism. An Iranian version of this measure was used, which was found to have an acceptable internal consistency with a Cronbach’s Alpha of 0.85 (Ghorbandordinejad, 2014).
Depression Anxiety Stress Scale-21 (Lovibond & Lovibond, 1996). This measure consists of 21 items and evaluates depression (e.g., I felt I wasn’t worth much as a person), anxiety (e.g., I was worried about situations in which I might panic and make a fool of myself), and stress (e.g., I tended to over-react to situations) with 7 items for each subscale. This measure uses a 4-point Likert scale from 1 (Did not apply to me at all) to 4 (Applied to me very much, or most of the time), with higher scores indicating a greater level of depression, anxiety, and stress. In this study, an Iranian version of this measure was used which was found to have an acceptable internal consistency with Cronbach Alphas ranging from 0.79 to 0.90 (Asghari et al., 2008).
Results
Face validity
Prior to administering the survey, an impact score index was used to estimate face validity, which measures the relevance, comprehension, and appropriateness of items from the participants' point of view. Eight married men and 8 married women stated their opinions about the items on a 5-point Likert scale from 1 (not important) to 5 (completely important). The impact score is calculated by multiplying the frequency of participants who chose the values of 4 or 5 by the mean score of each item. If the value of the impact score is greater than or equal to 1.5 (Hajizadeh & Asghari, 2011), the item has acceptable face validity. All items had values greater than 1.5, demonstrating acceptable face validity of all items.
Content validity
The Content Validity Index (CVI) and the Content Validity Ratio (CVR) were used to measure the content validity of the translated measure. The CVI estimates the simplicity, clarity, and relevancy of items from a group of experts' point of view. Eight psychologists were recruited to determine the content validity of the items. Experts evaluate the items on a scale from (1) not relevant at all to (4) highly relevant. The value of the CVI is calculated by dividing the number of experts who selected the values of 3 or 4 by the total number of experts. If the value of the CVI is greater than or equal to 0.7, the item has acceptable content validity (Cook & Beckman, 2006). The CVR estimates the essentiality of items from a group of experts' point of view. Experts evaluate the items by selecting one of the values from (1) not essential to (3) essential. The value of CVR is calculated subtracting the number of experts from the number of experts who selected a value of 3, then dividing this value by half the number of experts (i.e., N/2). Items have acceptable content validity if the value of CVR is greater than Lawshe's value (0.78) (Lawshe, 1975). As shown in Table 1, the CVI and CVR values indicated all items had acceptable content validity.
CVR and CVI for the items of Romantic Perfectionism Scale.
CVI and CVR values do not have significant levels
Data analyses
Preliminary data analysis to check for outliers and normality was conducted using the Statistical Package for the Social Sciences (SPSS version 24). No missing data was present as only surveys with complete responses were collected. Outliers were checked using a box plot, which demonstrated no outliers were present in the dataset. The data was found to be normally distributed as skewness (−0.87 to 0.38) and kurtosis values (−1.12 to 0.27) were in the acceptable ranges of ±2 and ±3, respectively (Tabachnick & Fidell, 2012).
Construct validity
A confirmatory factor analysis (CFA) in AMOS software (Version 24) was used to examine construct validity by testing hypothesized relationships between the observed variables (items) and the latent variables (factors). The CFA has three steps: Evaluation of factor loading values: If a factor loading value is less than 0.4, greater than 1, or negative, the item must be removed (Kline, 2015). As seen in Figure 1, the RPS with one-factor (Model A) and the RPS with third-order (Model B) with their factor loadings are presented. In Model A, the factor loading values of the DPB8 (0.39) and the DPB (0.26) are less than 0.4, while in Model B, all factor loadings of the items are above 0.4, therefore all items were kept in the analysis. Means, standard deviations, and item-to-total correlations of the included items were calculated (see Table 2). All correlation coefficients were acceptable indicating the homogeneity of the items in measuring the RPS. Evaluation of the measurement model fit indices: The following indices were used to examine model fit: CMIN/df < 5, Root Mean Squared Error of Approximation (RMSEA) < .08, Tucker- Lewis Index (TLI), Comparative Fit Index (CFI), and the Goodness of Fit Index (GFI) > .90 (Byrne, 2013). The results revealed Model B with two subscales had good fit indices (CMIN/df = 4.21, p < .01, RMSEA = .08, TLI = .92, CFI = .92, GFI = .91) compared to Model A with one-factor (CMIN/df = 8.71, p < .01, RMSEA = .16, TLI = .74, CFI = .81, GFI = .81). Evaluation of convergent validity and internal consistency among items of the measure. As seen in Table 3, the Average Variance Extracted (AVE) values were greater than 0.5, indicating acceptable convergent validity for the measure and the two subscales. Additionally, the Construct Reliability (CR) values were greater than 0.7, indicating the scale had acceptable internal consistency (Byrne, 2013).

The RPS construct Model A (one factor, first order) and the RPS construct Model B with two sub scales.
Means, standard deviations, and item-to-total correlations of the Romantic Perfectionism Scale.
Items with asterisk must be reverse coded.
AVE and CR for the sub scales of the Romantic Perfectionism Scale.
AVE: Average Variance Extracted; CR: Construct Reliability.
Concurrent validity
A correlation analysis was performed between the RPS and the other measures in order to assess concurrent validity. The results indicated that both of the two subscales of the RPS were significantly, positively related to discrepancy, standards, order, stress, anxiety, and depression (see Table 4).
Correlations between the studied variables.
**Significant at the 0.01 level.
Significant at the 0.05 level.
Invariance test of gender on the proposed model
A multi-group analysis was used to analyze the invariance test of gender in the proposed model ( Baron & Kenny, 1986). In the proposed model, gender could be considered a moderator if the variant model indices fit better than the invariant model. The results showed that the invariant model's measurement fit indices were better fit (CMIN/df = 5.21, p < .01, CFI = .83, RMSEA = .09, TLI = .85, GFI = .87) than the variant model (CMIN/df = 7.21, p < .01, CFI = .76, RMSEA = .16, TLI = .81, GFI = .82), suggesting no gender-based difference in the proposed model (Hair et al., 2006).
Discussion
The purpose of this study was to verify the psychometric properties of the RPS in a sample of Iranian couples. The results showed that the translated Persian version of the measure demonstrated satisfactory face and content validity. A confirmatory factor analysis supported the two-factor structure of the RPS (i.e., self-oriented romantic perfectionism and other-oriented romantic perfectionism) as was demonstrated in the original study conducted by Matte and Lafontaine (2012). All items remained in the scale and the two subscales demonstrated acceptable internal consistency and convergent validity. Internal consistency was demonstrated for the two subscales, and were comparable to the Cronbach Alpha values found in the original study (Matte & Lafontaine, 2012).
The results of the invariance analysis showed that the RPS did not differ significantly between men and women in the study, indicating that this scale could be used for individuals who identify with both genders. Finally, both subscales were positively associated with discrepancy, order, standards, depression, anxiety, and stress as measured by the Almost Perfect Scale-Revised and Depression Anxiety Stress Scale-21. In line with these findings, Stoeber et al. (2021) showed that self-oriented and other-oriented general perfectionism correlated with impatience and irritability, which can be harmful to interpersonal relationships.
There are several implications of this study given the Persian version of the RPS demonstrated satisfactory psychometric properties. As previous studies have shown, perfectionism is associated with relationship disatisfaction (Flett et al., 2001; Furman et al., 2017; Habke et al., 1999; Mackinnon et al., 2012), a valid and reliable measure of romantic perfectionism could help clinicians understand client expectations of themselves and their partner, and could be a useful addition to a battery of questionnaires used in Iranian couples' therapy.
Limitations
This study has a few limitations. A sample of Iranian couples in heterosexual marriages lasting more than one year was used to perform this study, so these findings may not be generalizable to non-heterosexual couples, couples who do not cohabitate, or newly married couples. This study also was performed on married men and women and may not apply to individuals who are in a relationship but not married. Future research should examine the psychometric properties of this scale on other populations. Test-retest reliability and age invariance were not performed in this study. It is recommended future research measures the test-retest reliability and age invariance. Researchers have hypothesized that perfectionism in the context of relationships might be a consequence of narcissism (Casale et al., 2016). Future studies could test this hypothesis in the context of marital relationships.
Conclusion
The findings of this study confirmed the two subscales of the Persian version of the Romantic Perfectionism Scale had acceptable validity and reliability, and can be used by clinicians and future researchers to assess romantic perfectionism in Iranian married couples.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the1964 Helsinki declaration and its later amendments or comparable ethical standards.
