Escala de Mitos do Amor Romântico (EMAR): Portuguese Version of the Scale of Myths of Romantic Love (SMRL)

Abstract

Background: Violence legitimization has been associated with the myths of romantic love. In the Portuguese context, there are no available standardized tools to assess it. This study aimed to develop the adaptation and validation of the Portuguese version of the Scale of Myths of Romantic Love (SMRL) in youth and emerging adults. Also, the acceptance of myths of romantic love was described. Methods: The sample comprised 165 individuals (83% female) aged from 18 to 30 (M = 23.01; SD = 3.33). Results: Confirmatory factor analysis supported the two-factor correlated structure found in the original validation: Idealized Love (IL) and Distorted Love (DL). The selected nested model obtained acceptable good-of-fit indices. Reliability analysis indicated acceptable internal consistency (α = 0.728; ω = 0.711) and the item-factor correlations were strong. Gender differences in the endorsement of romantic myths were found. Males obtained significantly greater acceptance of need for a couple and ambivalence myths. Females significantly endorsed the half orange myth. Conclusions: Our findings supported the two-factor related structure of the SMRL Portuguese version. Gender-based differences in romantic beliefs and their potential link to violence legitimization, gender stereotypes, and love perception are highlighted as in need of further studies. This research provides evidence regarding the psychometric properties of the Escala de Mitos do Amor Romântico (EMAR).

1. Introduction

In the past decade, the myths of romantic love have become an area of increasing interest in research involving young populations due to their association with violence.

Since the conceptualization of Sternberg’s (1986) triangular theory of love, the construct of love has evolved over time. New research trends have focused on understanding the mechanisms, ontogeny, functions, and phylogeny that explain romantic love (Bode and Kushnick 2021). In this regard, proximate explanations provide insight into the psychological processes and mechanisms that enable individuals to develop and experience romantic love (Bode and Kushnick 2021). Classical theoretical models of love, such as confluent love (Giddens 1992) and consummated love (Sternberg 1986), offer balanced and healthy views of love. Love perception in Portuguese culture is mainly passionate (eros—physical attraction and emotional intensity), committed (storge—companionship-driven love), and affectionate (agape—altruistic and selfless love) (Neto and Pinto 2025). Despite love being one of the most desirable emotions of human beings, the characteristics of romantic love promote believing in myths linked to this type of love (Jiménez-Picón et al. 2022). In this regard, the myths of romantic love are particularly relevant, as they represent how romantic love is perceived in intimate relationships, which is shaped by socio-cultural factors (Bonilla-Algovia and Rivas-Rivero 2018). From an anthropological point of view, myths are shared beliefs and narratives that form part of a culture’s collective consciousness (Bidney 1967). Understanding myths as a cultural phenomenon, they maintain a certain system of values and social norms (Shynkaruk et al. 2018). Specifically, the myths of romantic love were defined by Yela (2003) as a set of socially shared beliefs about the nature of love. Despite emerging theoretical models in recent years, there is a consensus in the scientific community regarding the following nine myths: omnipotence (love can overcome everything), half orange (there is a predestined partner for everyone), need for a couple (happiness cannot be achieved without a partner), eternal passion (the passion of infatuation must last over time), exclusiveness (one can only love one person), marriage (romantic love leads to marriage), abnegation (loving involves self-sacrifice for the partner), jealousy (jealousy is a sign of true love), and ambivalence (love and violence can coexist in a relationship) (Bonilla-Algovia and Rivas-Rivero 2020; Bosch et al. 2007; Cubells-Serra et al. 2021; Yela 2003). Previous research has suggested that some of these myths, such as jealousy and ambivalence, are associated with intimate partner violence (IPV) due to the negative influence these limiting beliefs have on the recognition of certain behaviors as forms of violence (Cava et al. 2020). In this way, the myths of romantic love contribute to the normalized perception of certain controlling behaviors (Linares et al. 2021) and are positively associated with the legitimization of IPV (Baldry and Pagliaro 2014; Hester et al. 2010; Lelaurain et al. 2021; Papp et al. 2017). Additionally, romantic love has been identified as a risk factor for gender-based violence (GbV) (Jiménez-Picón et al. 2022). Although several studies have analyzed romantic love myths in young populations across different countries and contexts (Cava et al. 2020; Fernández et al. 2023; Goenaga 2021; Rivas-Rivero and Bonilla-Algovia 2021; Standley 2022; Triano et al. 2021), it is noteworthy that there is a lack of empirical studies assessing romantic love myths in the Portuguese context.

According to the EU Gender-based Violence Survey (FRA et al. 2024), the prevalence of violence experienced by women aged 18–74 in the EU is 32%. The proportion is notably higher in the young (18–29 years) and young adult (30–44 years) populations, at 35% for both groups. In Portugal, the prevalence of physical and/or sexual gender-based violence is 10.3%. However, when psychological violence is included, this proportion rises to 22.5% (FRA et al. 2024). Regarding the legitimization of violence during dating, a recent national study (União de Mulheres Alternativa e Resposta [UMAR] 2024) demonstrated that young males legitimize certain types of violence more than females, including: control (64.1% vs. 46.8%), psychological violence (44.0% vs. 24.9%), stalking (39.9% vs. 24.0%), sexual violence (41.1% vs. 20.8%), social network violence (25.5% vs. 13.6%), and physical violence (13.9% vs. 6.3%). Also, several studies have obtained high rates of IPV in both adolescent (Neves 2016; Perista et al. 2012) and adult (Neves and Nogueira 2010) Portuguese populations.

Given the limited number of studies analyzing romantic love myths in young populations within the Portuguese context, as well as the alarming data on the legitimization of violence among the youth population in the country, there is a clear need for a valid and reliable psychometric instrument to assess the myths of romantic love in this context. In fact, the deconstruction of gendered narratives of romantic love has been emphasized as a necessity for over a decade in the Portuguese context (Neves and Nogueira 2010), but this gap still persists in terms of quantitative assessment. Tackling these research gaps requires a validated tool for evaluating romantic love myths. One of the most recent instruments for the assessment of the myths of romantic love is the Scale of Myths of Romantic Love (SMRL; Bonilla-Algovia and Rivas-Rivero 2020). This psychometric tool offers several advantages, as it enables the assessment of a wide range of romantic myths, including both idealized and distorted beliefs about love. Moreover, the SMRL allows for the assessment of the romantic myth of ambivalence, incorporating an implicit analysis of violence (Bonilla-Algovia et al. 2024). To the best of our knowledge, there is currently no validated instrument in Portuguese to assess the myths of romantic love.

Therefore, this study aimed to adapt and validate the Scale of Myths of Romantic Love (SMRL) into Portuguese. The hypothesis of this study was that the two-factor correlated structure of the original test would be confirmed in a sample of Portuguese youth and young adults.

2. Materials and Methods

2.1. Participants

The sample consisted of 165 young people from Portugal, aged between 18 and 30 years (M = 23.01; SD = 3.33). Most participants were female (83.0%). It is noteworthy that both the age mean and the gender distribution were similar to those in the original validation (Bonilla-Algovia and Rivas-Rivero 2020). Participants were recruited using a non-probabilistic snowball sampling method (Goodman 1961) via social media. The sample comprised youth and young adults from the general population, including university students enrolled in Psychology, Communication Studies, Translation Studies, and Social Work programs. Inclusion criteria were as follows: aged between 18 and 30 years old, Portuguese nationality, and residence in Portugal. It is important to mention that the sample size and the proportion of male participants in this study represent a significant limitation.

2.2. Measurements

Myths of romantic love. The Scale of Myths of Romantic Love (SMRL; Bonilla-Algovia and Rivas-Rivero 2020) is a self-report questionnaire comprising 11 items rated on a five-point Likert scale (1 = Completely disagree, 2= Disagree, 3 = Neither agree nor disagree, 4 = Agree, 5 = Completely agree). The SMRL assesses the acceptance of the following myths of romantic love: omnipotence (item 1), half orange (items 2 and 3), need for a couple (item 4), eternal passion (item 5), exclusiveness (item 6), marriage (item 7), abnegation (items 8 and 9), jealousy (item 10), and ambivalence (item 11). In terms of construct validity, the SMRL is comprised of a two-factor correlated internal structure: Idealized Love (Factor 1. IL) and Distorted Love (Factor 2. DL). Moreover, according to the original authors, the SMRL also allows for the assessment through the total score on the scale. With regard to validity and reliability, the goodness of the fit indices tested (RMSEA, SRMR, GFI, AGFI, CFI, and AIC) were excellent. The internal consistency of SMRL is acceptable according to cutoff criteria (George and Mallery 2018), as Cronbach’s alpha (α = 0.73) is higher than 0.70.

Sociodemographic variables. Age, sex, and gender data were also collected.

2.3. Procedure

Data were self-reported and collected via the SurveyMonkey online platform (SurveyMonkey Inc. n.d.). Permission for translation and adaptation of the instrument into Portuguese was first obtained from the original authors. The SMRL adaptation process followed a back-translation procedure, consistent with established guidelines for the translation and adaptation of psychological assessments (Hambleton and De Jong 2003; International Test Commission [ITC] 2017). In the first step, a native Portuguese-speaking translator translated the original Spanish version of the SMRL into Portuguese. Subsequently, a second translator, fluent in both languages, back-translated the Portuguese version into Spanish. The research team then compared the original Spanish version with the back-translated version to ensure both semantic and conceptual equivalence.

2.4. Data Analysis

Data analyses were performed using IBM SPSS Statistics v. 29 (IBM Corporation 2023a) and IBM SPSS AMOS v. 29 (IBM Corporation 2023b). SPSS Statistics was employed to assess the normality of the data distribution, examine descriptive statistics for the items, and conduct confirmatory factor analysis (CFA). The CFA was used to evaluate the two-factor correlated model proposed by the original authors. Based on the original validation of the SMRL, where the two factors were allowed to correlate (Bonilla-Algovia and Rivas-Rivero 2020), the CFA was performed using oblimin oblique rotation with Kaiser normalization, applying both maximum likelihood and robust estimation methods. To assess internal consistency, both Cronbach’s alpha and McDonald’s omega were computed, as McNeish (2018) highlighted the limitations of Cronbach’s alpha. Additionally, independent group comparisons were conducted using the Mann–Whitney U test to examine potential gender differences in both factors of SMRL. AMOS Statistics was used to determine the goodness-of-fit of the obtained model. Additionally, the one-factor and the two-factor uncorrelated models were also tested, in the same way as the original SMRL validation. The following goodness-of-fit indices were considered: the normed Chi-square statistic (CMIN/df), the Jöreskog–Sörbom Goodness of Fit Index (GFI), the Steiger–Lind Root Mean Square Error of Approximation (RMSEA), the Parsimonious Normed Fit Index (PNFI), the Parsimony ratio (PRATIO), and the Akaike Information Criterion (AIC). Values of the CMIN/df index between 2 and 3 are indicative of an acceptable model fit (Kline 2023). Values of the GFI higher than 0.90 imply acceptable fit (Bentler 1990; Hu and Bentler 1999). RMSEA values lower than 0.10 are considered indicative of acceptable fit (Marcoulides and Yuan 2017) and reasonable error (MacCallum et al. 1996). Values of PNFI and PRATIO higher than 0.50 are indicative of good parsimony of the model (Mulaik et al. 1989). Also, the AIC allows to select the best model in terms of both parsimony and fit adequacy of the data (Akaike 1974), as the model with the lowest AIC value is considered the best (Brown 2015).

2.5. Ethical Approval

This research obtained ethical approval from the Ramon Llull University Ethical Committee (Code 2122013D, 31 May 2022).

3. Results

3.1. Descriptive Analysis of Items

Table 1. Descriptive statistics of EMAR items.

Item	M (SD)	%	Skewness	Kurtosis
1. True love can overcome any difficulty or problem [O amor verdadeiro pode superar qualquer dificuldade ou problema]	3.47 (1.19)	58.2	−0.393	−0.960
2. Everyone will find their soulmate at some point [Todas as pessoas encontrarão a sua alma gémea num determinado momento]	2.99 (1.15)	35.1	0.048	−0.856
3. When the perfect partner is found, they should not be let go [Quando se encontra o parceiro perfeito, há que não o deixar escapar]	3.25 (1.10)	41.8	−0.064	−0.764
4. A person will not be fully happy until he or she finds a partner who complements him or her [Uma pessoa não será plenamente feliz até que encontre um parceiro que a complete]	2.20 (1.16)	15.7	0.790	−0.232
5. It is only true love when the passion of the first months is maintained [Só é amor verdadeiro quando a paixão dos primeiros meses se mantém]	2.15 (1.02)	13.3	0.767	−0.067
6. You cannot love more than one person at the same time [Não se pode amar mais do que uma pessoa ao mesmo tempo]	3.26 (1.27)	43.6	−0.067	−1.125
7. True love must end in marriage [O amor verdadeiro deve terminar em casamento]	2.52 (1.20)	24.2	0.298	−0.959
8. A person in love cares more for the well-being of his or her partner than for his or her own [Uma pessoa apaixonada preocupa-se mais com o bem-estar do seu parceiro do que com o seu próprio]	2.88 (1.17)	35.2	0.169	−1.016
9. To love someone is to do everything for that person, even if it sometimes means doing things you do not like [Amar alguém é fazer tudo por essa pessoa, mesmo que por vezes isso signifique fazer coisas de que não gostas]	3.26 (1.26)	49.7	−0.192	−1.142
10. Jealousy is a demonstration of love for one’s partner [O ciúme é uma demonstração de amor pelo parceiro]	2.32 (1.12)	18.8	0.312	−1.073
11. Shouting and insults from a partner can be forgiven as long as there is true love [Os gritos e os insultos de um parceiro podem ser perdoados se existir amor verdadeiro]	1.87 (0.99)	8.5	0.910	−0.091

Note. M = Mean; SD = Standard Deviation; % = Proportion of participants reporting “agree” and “completely agree”. The Portuguese translation of the items is presented in square brackets.

presents the descriptive statistics for the items of the EMAR scale. As for the endorsement of myths of romantic love, the omnipotence (item 1, 58%), abnegation (item 9, 50%), exclusiveness (item 6, 44%), and half orange (item 3, 42%) myths were the most accepted by Portuguese youth and emerging adults. In contrast, item 11, which assesses the ambivalence love myth, presented the lowest endorsement. Similar mean values were observed across the rest of the items. Notably, items 4 and 11 exhibited higher levels of skewness compared to the other items within the same factor. Consistently, the kurtosis values for these two items were significantly lower than those for the remaining items. The sample-to-item ratio was 15:1, which is considered excellent (Kline 2023; Schumacker and Lomax 2015).

3.2. Internal Structure

The factor loadings are presented in

Table 2. Factor loadings and communalities from the CFA of EMAR.

Item	Factor Loadings		Communalities
	F1. Idealized Love	F2. Distorted Love
Item 1	0.434	—	0.204
Item 2	0.742	—	0.682
Item 3	0.535	—	0.292
Item 4	0.451	—	0.217
Item 5	0.395	—	0.170
Item 6	0.398	—	0.180
Item 7	0.330	—	0.290
Item 8	—	0.424	0.245
Item 9	—	0.665	0.466
Item 10	—	0.318	0.305
Item 11	—	0.471	0.236

. Items 1 to 7 loaded onto the first factor, Idealized Love (IL), while items 8 to 11 loaded onto the second factor, Distorted Love (DL). The standardized factor loadings for the IL factor ranged from 0.330 to 0.742, and for the DL factor, they ranged from 0.318 to 0.665 (see Table 2). Communalities ranged from 0.170 to 0.682, indicating a moderate to strong level of variance explained by the factors for the majority of the items.

Thus, the CFA outcomes supported the two-factor correlated structure proposed by the original validation (Bonilla-Algovia and Rivas-Rivero 2020), with a 41.50% explained variance. The fit of three models was tested (see

Table 3. Goodness-of-fit indices of the tested models.

	χ2	df	χ2/df	GFI	RMSEA [CI 95%]		PNFI	PRATIO	AIC
M1	90.747	43	2.110	0.907	0.082	[0.059–0.106]	0.552	0.782	136.747
M2	113.527	44	2.580	0.881	0.098	[0.076–0.121]	0.506	0.800	157.527
M3	114.310	44	2.598	0.892	0.099	[0.077–0.121]	0.504	0.800	158.310

Note. M1 (model 1) is the two-factor correlated structure model proposed by the original authors. M2 (model 2) has a uni-factorial structure. M3 (model 3) is a two-factor uncorrelated structure model. Cutoff points are described in the data analysis of the Materials and Methods Section.

). Models 2 and 3 showed a poor fit, with unsatisfactory goodness-of-fit indices. Therefore, Model 1 (see Figure 1) obtained an adequate fit according to cutoff values (χ²(43) = 90.747, p < 0.001, χ²/df = 2.110, GFI = 0.907, RMSEA = 0.082, PNFI = 0.552, PRATIO = 0.782). Also, the AIC index suggested that Model 1 was the best (Δχ² = −22.78; Δdf = −1; Δχ²/df = −0.47; ΔAIC = −20.78). The two EMAR factors correlated positively and statistically significantly (r = 0.383***; R² = 0.147), which reinforced the maximum likelihood extraction method applied.

3.3. Internal Consistency

Internal consistency was assessed using reliability coefficients and item-factor correlations (see

Table 4. Item-factor Spearman’s correlations.

	F1. Idealized Love (IL)	F2. Distorted Love (DL)
Item 1	0.629 ***	—
Item 2	0.588 ***	—
Item 3	0.656 ***	—
Item 4	0.594 ***	—
Item 5	0.472 ***	—
Item 6	0.593 ***	—
Item 7	0.515 ***	—
Item 8	—	0.674 ***
Item 9	—	0.771 ***
Item 10	—	0.662 ***
Item 11	—	0.546 ***

Note. *** p ≤ 0.001.

). The results of the Cronbach’s alpha and McDonald’s omega analyses indicated acceptable internal consistency (α = 0.728; ω = 0.711) (George and Mallery 2018). Additionally, item analyses revealed that all items contributed to the internal consistency of both factors, as the removal of none of them resulted in an increase in reliability. Item-factor correlations were statistically significant, with values approaching 0.5, further supporting the adequacy of internal consistency based on both reliability coefficients and item-factor correlations.

3.4. Between-Group Gender Differences in Myths of Romantic Love

Regarding the comparison between females and males (see

Table 5. Between-group Mann–Whitney U test comparison by gender (N = 165).

Item (Myth)	Gender
	Female (n = 137)		Male (n = 28)
	M (SD)	%	M (SD)	%	MD	U	rb	p
Item 1 (omnipotence)	3.50 (1.20)	58.39	3.32 (1.16)	57.14	0.18	1739.00	0.093	0.419
Item 2 (half orange)	3.07 * (1.15)	37.96	2.57 * (1.07)	21.43	0.50	1445.50	0.246	0.034
Item 3 (half orange)	3.25 (1.14)	41.61	3.25 (0.89)	42.86	0.00	1893.50	−0.013	0.912
Item 4 (need for a couple)	2.12 ** (1.17)	14.60	2.61 ** (1.03)	21.43	−0.49	1373.50	−0.284	0.014
Item 5 (eternal passion)	2.18 (1.07)	15.33	2.00 (0.77)	3.57	0.18	1815.00	0.054	0.636
Item 6 (exclusiveness)	3.30 (1.27)	45.26	3.07 (1.25)	35.71	0.23	1714.00	0.106	0.364
Item 7 (marriage)	2.49 (1.24)	24.82	2.64 (0.99)	21.43	−0.15	1728.00	−0.099	0.396
Item 8 (abnegation)	2.85 (1.18)	33.58	3.04 (1.11)	42.86	−0.19	1743.50	−0.091	0.432
Item 9 (abnegation)	3.18 (1.29)	45.99	3.68 (1.06)	67.86	−0.50	1502.00	−0.217	0.063
Item 10 (jealousy)	2.31 (1.10)	17.52	2.39 (1.23)	25.00	−0.08	1862.50	−0.029	0.803
Item 11 (ambivalence)	1.80 * (0.96)	7.30	2.25 * (1.08)	14.29	−0.45	1448.50	−0.245	0.029
F1. Idealized Love	2.85 (0.70)	—	2.78 (0.53)	—	0.07	1823.00	0.050	0.679
F2. Distorted Love	2.53 (0.76)	—	2.84 (0.77)	—	−0.31	1497.00	−0.219	0.066

Note. M = Mean; SD = Standard Deviation; % = Proportion of participants reporting “agree” and “completely agree”; MD = Mean Difference (negative values indicate higher mean in the male group); U = Mann–Whitney’s statistic value; r_b = Rank-biserial correlation indicating the effect size (Cohen 1992): r_b ≈ 0.1 (small); r_b ≈ 0.3 (medium); r_b ≈ 0.5 (large); p= probability value: * p ≤ 0.05, ** p ≤ 0.01.

), the Mann–Whitney U test indicated no significant differences (p > 0.05) for the Idealized Love (IL) or Distorted Love (DL) factors. However, item-level analyses revealed statistically significant differences for items 2, 4, and 11 between females and males (p < 0.05). Specifically, item 2 had significantly higher results for the female group, indicating greater acceptance of the half orange myth (38% vs. 21%). In contrast, the need for a couple (15% vs. 21%) and ambivalence (7% vs. 14%) myths had significantly higher results for the male group. Rank-biserial correlation (Cohen 1992) indicated a medium effect size (r_b ≈ 0.3) for the obtained between-group differences.

4. Discussion

This research analyzed the psychometric properties of the Portuguese version of the Scale of Myths of Romantic Love (SMRL) in a Portuguese sample of youth and emerging adults. Moreover, it aimed to explore the endorsement of myths of romantic love.

First, our findings generally support our methodological research hypothesis, as confirmatory factor analyses supported the two-factor correlated structure of the original test (Bonilla-Algovia and Rivas-Rivero 2020). Thus, the EMAR is comprised of two factors: Idealized Love (F1. IL) and Distorted Love (F2. DL). The findings indicated adequate item properties and internal consistency.

Second, the results of this study have suggested that the myths of romantic love are quite frequent among the Portuguese youth population. Specifically, the omnipotence, half orange, exclusiveness, and abnegation myths presented scores above the middle point of the Likert Scale, indicating a tendency towards agreement and acceptance. These results are in line with recent studies obtained in the Spanish context (Bonilla-Algovia and Rivas-Rivero 2020; Bonilla-Algovia et al. 2024).

Third, as for gender differences, this research has suggested potential differences between males and females in the half orange (item 2), need for a couple (item 4), and ambivalence (item 11) myths. As can be seen in Table 5, a statistically significant greater acceptance of the half orange myth was reported by the female group, whereas the male group obtained significantly higher scores in the need for a couple and ambivalence romantic love myths (p < 0.05). These outcomes contradict previous research that has obtained similar perceptions between both genders (Cubells-Serra et al. 2021; Ferrer et al. 2010). This discrepancy may be attributed to several factors: sample size, age range (e.g., 18–70 y.o., Cubells-Serra et al. 2021; 18–93 y.o., Ferrer et al. 2010), and/or potential cultural differences in love interpretation (Karandashev 2017), as both studies included Hispanic samples. Along these lines, previous cross-cultural studies on romantic love found that love models vary more between cultures than between genders (De Munck et al. 2009; Nelson and Yon 2019). Nelson and Yon (2019) found some exceptions to this pattern, as they identified some gendered patterns in responses across different cultures, predominantly in the negative or irrational aspects of romantic love. A plausible hypothesis to explain this contradiction is that the adult population shows greater acceptance of romantic love myths compared to the young adult population. In this sense, a recent study conducted in Portugal (Cipriano et al. 2021) demonstrated that traditional gender stereotypes (such as the association of strength with men and nurturing and passivity with women) are more strongly accepted by older individuals and women. Therefore, it is hypothesized that certain stereotypes (e.g., the expectation of women being caring and passive) could be linked to the abnegation myth of romantic love, which involves prioritizing a partner’s needs over one’s own. Furthermore, the stereotypical view of men as strong and dominant can be associated with the myth of ambivalence, often reflected in aggressive behaviors like shouting or insults. Also, controlling behaviors are stereotypically linked to men, which could be related to the myth of jealousy. These hypotheses suggest an interrelation between gender stereotypes and myths of romantic love, implying that certain gendered expectations may influence the way love is perceived in the context of partner selection.

4.1. Strengths

It is noteworthy that the translation and back-translation process adhered to the guidelines of the International Test Commission [ITC] (2017). Moreover, the sample in the present study exhibited the same age range, mean age, and gender distribution as those in the original validation, ensuring the reproducibility of the original study and maintaining the same sample characteristics. Additionally, the reliability results and model fit indices for the bi-factorial model are similar to those obtained in the original study. For example, in this study, the obtained Cronbach’s alpha (α = 0.728) is nearly the same value as the original validation (α = 0.73; Bonilla-Algovia and Rivas-Rivero 2020). These similarities indicate consistency in our psychometric findings.

4.2. Limitations and Future Directions

This study is not without limitations. Despite the excellent person-to-item ratio (Schumacker and Lomax 2015), it is noteworthy that the sample size obtained is not large enough to generalize the gender between-group differences found in the population of young and young-adult people in Portugal. Thus, the sample size and gender distribution limit the generalizability of our between-group gender comparison findings. Additionally, the snowball sampling method involves a potential selection bias (Parker et al. 2019). Thus, to expand the generalizability of future studies, it is recommended to use randomly selected samples from a cross-section of Portuguese youth and young adults. On the other hand, based on the premise that romantic love exists across the human lifespan in both sexes (Bode and Kushnick 2021), as far as we know, there is no extensive literature on myths of romantic love across generations. Therefore, future research involving a broader age range is needed to analyze potential differences in the representation of love across adolescent, adult, and older age groups, while also further examining the intersectionality of age and gender in relation to gender stereotypes and myths of love. From a statistical perspective, a notable limitation is the absence of concurrent validity analysis, as there is no other validated psychometric measure in Portuguese that assesses the same construct nor another construct that could be considered a gold standard. However, the lack of validated instruments for concurrent validity is a common methodological issue in the field of psychometric instrument development and adaptation (DeVellis and Thorpe 2021).

Future research could benefit from the EMAR Portuguese validation and use it as a measure of concurrent or convergent external validity in the development of alternative psychometric tools for assessing perceptions of romantic love and related myths. Furthermore, to investigate the relationship between the acceptance of romantic love myths and violence, future studies could adopt a longitudinal design to explore the predictive capacity of romantic love myths on violence perpetrated and/or suffered in young populations. Also, this scale could be valuable for the assessment of myths of romantic love endorsement across different cultures. In this regard, it could be hypothesized that countries with less idealized and distorted attitudes towards love will present less interpersonal violence.

Additionally, given the significant impact and detrimental effects of gender stereotypes on women (Banchefsky and Park 2018), future research should focus on examining the connections between specific gender stereotypes and myths of romantic love. Finally, additional relevant variables, such as love perception and relationship quality (consensus, satisfaction, cohesion, and affective expression), could be analyzed in conjunction with myths of romantic love using Sternberg’s Triangular Love Scale (STLS; Cassepp-Borges and Pasquali 2012) or the Love Attitudes Scale, Short Form (LAS-SF; Neto and Pinto 2025) and the Dyadic Adjustment Scale (DAS; Gomez and Leal 2008).

5. Conclusions

Overall, the results show that the Portuguese version of the SMRL (Escala de Mitos do Amor Romântico, EMAR, see Appendix A) is an appropriate instrument for evaluating perceptions of love, demonstrating robust psychometric properties. One of the main benefits of offering the Portuguese-language version of the SMRL test is that it enhances the ability to make comparisons across different samples and ensures future reproducibility of these findings.

Also, from a primary prevention perspective, due to the beliefs that shape romantic love myths increasing the acceptance of behaviors that would typically be considered abusive in a relationship (Lelaurain et al. 2021), the EMAR can be very useful for early detection of myths that may lead to dating violence and IPV. In fact, a recent longitudinal study (Neves et al. 2021) has demonstrated that traditional marital beliefs in adolescence are predictors of dating violence in young adulthood. Therefore, the relationship between romantic love myths and violence appears to be an emerging field of study in the Portuguese context.

To sum up, the assessment of romantic love myths is crucial to recognizing their contribution to the dynamics, communication, and conflicts emerging in intimate relationships. We hope that this study will serve as a psychometric foundation for future research on the myths of romantic love within the Portuguese context, contributing to the generation of essential knowledge on this dyadic phenomenon.