Reported father involvement and indicators of subjective well-being in transition to adulthood

ABSTRACT

This study examined the possibility of predicting the indicators of adolescents’ subjective well-being (satisfaction with life and frequency of positive and negative experiencing) based on the reported father involvement, which was the reason for verifying the factor structure and examine the psychometric properties of the Slovak version of the Reported Father Involvement Scale (R-FIS) in adolescents. The research sample consisted of 400 students (290 girls), aged 15 to 18 years (M = 16.51, SD = 1.06). Regarding the Slovak version of the R-FIS, the results of the reliability (McDonald’s ω) estimation were acceptable. The CFA (using polychoric correlations, DWLS method) results supported the 1-factor structure of the scale, which was used in further analyses. The results of the analyses confirmed our expectations, since the father involvement, when controlling for the intervening variables (gender, age, and cohabitation with the father), proved to be one of the significant predictors of subjective well-being indicators.

KEYWORDS:

• Father involvement
• subjective well-being
• adolescents
• the Reported father involvement scale

Introduction

The father becomes an important part of the child’s life from the moment they are born. The results of various current studies (e.g. Cabrera et al., 2018; Henry et al., 2020) confirm that parents play a very important role in a child’s biological, psychological, emotional, and social development, especially in childhood. The relationship with the parents naturally changes during the child’s life. Adolescence is a period peculiar for the initiation of the individuation process; notwithstanding, parents’ importance in the social environment of adolescents remains (Laursen & Collins, 2009; Smetana & Rote, 2019). The important role of family relationships is also confirmed by the results of research (Lacinová et al., 2018; Mónaco et al., 2019) that focused on comparing the importance of incoming peer and romantic relationships with relationships with parents. Likewise, the relationship with the father undergoes significant changes during this stage of life, mainly due to the gradual acquisition of greater autonomy and independence of the adolescent (Henry et al., 2020). In this context, Macek (2003) emphasizes that the relationships and bonds of adolescents with their fathers remain an influential factor of socialization and individualization during the entire stage of adolescence, as in the previous stages of development. Current research trends (e.g. Trahan et al., 2021) highlight the need to maintain a close relationship with the father during this stage, as it continues to support the healthy psychological and physical development of adolescents and is significantly related to better educational outcomes (Curtis et al., 2017; Whitney et al., 2017). According to Little et al. (2019), a close relationship with the father can also be a prevention against risky behaviour of adolescents (e.g. abuse of psychoactive substances, rioting, etc.).

In general, the role of the father was mainly associated with material and financial security in the past (Finley & Schwartz, 2004; Lamb, 2010). A change in the perception of the father’s functions has begun to gradually occur since the end of the last century, when the question of his involvement in raising a child came into focus (Finley & Schwartz, 2004; Hawkins & Palkovitz, 1999; Lamb, 2010). Lamb et al. (1985) were the first to systematically examine the father involvement in parenting and defined it as the time a father spends with his child in direct interaction. At the same time, they considered the father involvement in relation to three main aspects: the extent of interaction, availability, and the degree of responsibility. Hawkins and Palkovitz (1999) expand the definition of the term and emphasize that it is more than just spending time with the child and father’s physical closeness. According to them, involvement includes the father involvement in various areas of the adolescent’s life. They describe specifically three dimensions in this context: cognitive, affective, and behavioural, therefore they consider it difficult to observe the father involvement directly. This way of defining the father involvement is currently supported, for example, by Su et al. (2017) in their research. A more recent reconceptualization of the construct has been proposed by Pleck (2010). He considers three main components of the father involvement – positive activities, kindness, responsiveness and control, and two minor ones – indirect care and process responsibility. Based on the literature review, Babinčák et al. (2020) add the most essential tasks of a father regarding his involvement, which is being strict and demanding discipline as well as being kind and loving. Finley and Schwartz (2004), authors of the Father Involvement Scale, based on both the theory of Hawkins and Palkovitz (1999) and their own research, distinguish three main categories of the father involvement. In the first, which is represented by expressive involvement, the authors include, for instance, spending time together, emotional, social, physical, and spiritual development, care, or engaging in games and entertainment during free time. The second category, called instrumental involvement, includes, according to the authors, the following areas in which the adolescent evaluates the father involvement: discipline, protection, providing income, developing responsibility and independence. The third category, which the authors identify as mentoring, refers to areas such as education, competence development, advice provision, and intellectual development. It is this category that became discussed during the verification of the psychometric properties of the proposed scale (Finley et al., 2008), considering that it represents a conceptual and empirical overlap between instrumental and expressive involvement of the father. In this context, studies are also currently emerging in which the father involvement represents one comprehensive construct (e.g. Alifa & Handayani, 2021; Cracco et al., 2022).

Based on the results of a recent meta-analysis that included 34 studies (Jeynes, 2016), we can conclude that fathers play a unique role in children’s development that differs from that of mothers, and this applies to both boys and girls in different age cohorts. Despite the confirmation of this fact, some studies devoted to the involvement of the father in the context of gender (e.g. Morman & Floyd, 2006) state that the father is more involved in raising boys than girls. However, Lee (2020) points out that this difference narrows over time and the importance of the father in the lives of girls increases with age. The results of some research studies (e.g. Sağkal et al., 2018) confirm that those girls who perceive their father as warm, supportive, and involved have higher self-esteem and better psychological health. For boys, the father still remains important, especially as a model of behaviour. The findings of Pitsoane and Gasa (2018) suggest that the father involvement in the life of adolescent boys can also serve as a prevention against risky behaviour, since it promotes positive adaptation and emotional stability in boys.

The question of direct links between the father involvement and the subjective well-being of adolescents has not yet been adequately answered. Subjective well-being is defined through two components (Džuka et al., 2021) based on which people evaluate their experiencing (Diener, 2000), namely affective (frequency of experiencing positive and negative emotions) and cognitive (evaluation of general life satisfaction) (Džuka & Dalbertová, 2002). The results of research studies regarding the father involvement often focus indirectly on possible factors affecting subjective well-being. Several authors (Piko & Hamvai, 2010; Smetana & Rote, 2019; Williams & Kelly, 2005) generally state that the father involvement is significant when it comes to the psychological aspects of adolescent boys’ and girls’ lives, for instance, motivation, subjective well-being, social relationships (e.g. with partners, extended family members). The father involvement improves adaptation and healthy psychosocial adjustment (Jabeen et al., 2013), emotional, behavioural, and educational outcomes in adolescents (Cabrera et al., 2018). It increases the self-esteem and self-confidence of adolescents and facilitates the transition to adulthood (Su et al., 2017). Neither the negative effect of the father’s absence in the family on the experiencing of negative emotions nor the impact of living with the father in the family on the experiencing of subjective well-being was recorded (Flouri, 2006); however, the relationship between the closeness of the relationship and the father involvement in raising a child with the experiencing of subjective well-being was proven (Booth et al., 2010). Despite the fact that the experiencing of subjective well-being is influenced by several factors, Trahan et al. (2021) emphasize the role of a positive relationship with parents during adolescence stage in life as a significant predictor of subjective well-being. Piko and Hamvai (2010) point to gender differences and state that spending time together and communicating with parents increases subjective well-being in adolescent girls, but in boys parental support seems important. In their earlier review study, Rohner and Veneziano (2001) note the strong influence of paternal love on the social, emotional and cognitive development and functioning of children and young adults. The cited literature review draws attention to the significant influence of the father’s love on the further development of the child. In some cases, it can be more significant than the effect of the mother’s love. This notion complements (or even changes) the traditionally accepted unique role of the mother in upbringing. Specifically, for instance, Sağkal et al. (2018) find that a positive relationship with the father is associated with higher satisfaction of adolescent girls’ psychological needs and experiencing subjective well-being. Pitsoane and Gasa (2018) state that in the case of boys, the father low involvement in their upbringing is associated with the experiencing of negative emotions and a reduced level of subjective well-being.

In order to bring the cultural background closer, we must state at the end of the theoretical introduction that research on the father’s involvement is not widespread in Slovakia. In recent years, only a few articles focused on various aspects of fatherhood have been published. For example, Janoško and Janušková (2022) in their research focused on the views of contemporary fathers on the role of the father in the family confirmed the diverse forms of contemporary fatherhood. The authors identified fathers who are less involved in the family, but also those who put the family first. They confirmed the increase in the number of men who are very active in family life and in the lives of their children and who no longer consider supporting the family as their main role but see fatherhood as a mission to take care of children. Šeboková (2018) investigated the connection between the father’s involvement and the subjective well-being of Slovak university students. Indicators of well-being were related to adolescent reports, but not to fathers’ reports, of father involvement. Thus, the influence of fathers on adolescents is more precisely manifested in how children perceive his involvement than in his actual involvement. Fucsková et al. (2022) are the authors of the theoretical study focused on an overview of methods for assessing the relationship between father and child in early childhood, while the authors emphasize the need for validation of existing methods – also within Slovak culture. The topic of fatherhood is far from exhausted, which was one of the reasons why we focused on the father and his involvement.

On the basis of the aforementioned theoretical framework and due to the need to address an unexplored topic in Slovakia, the main aim of the presented study is to verify the predictive power of the father involvement regarding the affective and cognitive components of the adolescents’ subjective well-being. We assume that the higher level of the father involvement reported by adolescents is positively related to subjective well-being. Specifically, we expect that it increases the frequency of positive experiencing and the level of general life satisfaction and, on the contrary, decreases the frequency of negative experiencing. In the context of this aim, the first step is also to pilot test the Slovak translation of the questionnaire by Finley and Schwartz (2004) that determines the degree of the father involvement in the life of adolescents and was used in the presented research.

Method

Research sample

The research sample consisted of 442 secondary school students from all over Slovakia, out of which 42 respondents were excluded for reasons of not meeting the age criteria, not giving consent to data processing, and incorrectly filling out the questionnaire. Age of respondents was from 15 to 18 years (M = 16.51, SD = 1.06). The resulting group consisted of 110 (27.5%; age: M = 16.6, SD = 0.99) boys and 290 (72.5%, age: M = 16.5, SD = 1.09) girls. Of the total number of respondents (400), 293 respondents stated that they lived in the same household with their father (of which 206 were girls and 87 were boys). Data collection took place online during the months of November and December 2020. The approached schools were selected from the register of secondary schools in Slovakia via a random quota selection based on the region. To ensure the sample representativeness, schools from all regions were approached (both state and private gymnasiums (grammar school), secondary professional schools, secondary vocational schools). The involvement of the respondents in the research was mediated by the schools that agreed to participate in the research. The students were selected based on the age criterion.

Instruments

An informed consent was obtained prior to the administration of the questionnaire battery. It contained information about the purpose of the research, the use of data, and voluntary and anonymous participation in the research. Given that the respondents were over 15 years old, the informed consent was obtained directly from them, which is in compliance with Slovak legislation. Disagreement with participation was accepted and respondents (4) were excluded from the research. To maintain the respondent’s attention, we included a control question in the test battery, formulated as follows: „In the following question, please mark the number 4.” The battery of questionnaires consisted of a socio-demographic questionnaire ascertaining age, gender, cohabitation with the father in a joint household, and the following methodologies:

Subjective Well-Being

The Satisfaction with Life Scale (SWLS, Dalbertová, 1992; author of the Slovak translation is Džuka & Dalbertová, 2002) was used to measure general satisfaction with life as a cognitive component of subjective well-being. It is one of the two scales of the original instrument (Dalbertová, 1992), which consists of seven items, for instance, ‘My life could hardly be happier than it is.’ Respondents rated the items on a 6-point scale (1 = does not fit at all; 6 = fits completely), and a total score was evaluated (a higher value represented a higher level of life satisfaction).

The Scales of Emotional Habitual Subjective Well-being (SEHW, Džuka & Dalbertová, 2002) were used to determine the emotional component of subjective well-being. Of the total number of ten items, four items focus on the frequency of positive experiencing (pleasure, physical freshness, joy, happiness) and six items on the frequency of negative experiencing (feelings of guilt, shame, fear, pain, sadness, anger). Respondents in the questionnaire indicated how often they experienced the listed conditions on a scale of 1 = almost never to 6 = almost always. When evaluating the scale, the higher an individual scores, the higher the frequency of positive and negative experiencing is.

Father involvement

The Reported Father Involvement Scale (R-FIS) is one of the two subscales of the Father Involvement Scale (Finley & Schwartz, 2004) and was used to determine the level of the father involvement reported by adolescents. The original English version of the scale was translated into Slovak using the back translation method. The comprehensibility of the methodology was verified before the actual distribution through administration in a focus group of adolescents. The instrument consists of twenty items saturating two factors. Ten items make up the factor called Expressive Father Involvement, namely in the following areas: intellectual, emotional, social, spiritual and physical development, leisure/entertainment/games, sharing activities/interests, care, advise, friendly relationship. Ten items make up the second factor called Instrumental Father Involvement and relate to areas such as ethical/moral development, responsibility development, independence, competencies, providing income, cognition/learning, protection, discipline, school/homework. Following the administration of the scale in the focus group, which consisted of seven respondents aged 15–18 years, item no. 6 concerning career development was found irrelevant and subsequently excluded. The task of the respondents was to indicate how much their father participates in individual aspects of their life and development on a scale of 1 = never participates to 5 = always participates, with the higher the score the respondents achieve, the higher the level of the father involvement they perceive.

Data analysis

The data were processed using the programs Jamovi 1.6.15, Jasp 0.15.0.0, and RStudio. The dependent variables were indicators of subjective well-being (general life satisfaction, positive and negative experiencing); the independent variable was the reported father involvement. The relationship between the variables gender and cohabitation with the father in the same household and other variables was determined by means of the point-biserial correlation coefficient due to the nature of these variables. Pearson’s correlation coefficient was used to examine the relationships between age, dependent and independent variables.

At first, because standardized measures for the Slovak version of the Reported Father Involvement Scale (Finley & Schwartz, 2004) are not currently available to researchers the factor structure of Slovak version was verified via confirmatory factor analysis (CFA) using polychoric correlations (DWLS method). A similar approach was followed, for example Shafer et al. (2019). In the literature, three-factor solutions are presented. Based on the results of the Confirmatory Factor Analysis (CFA), the version that offers the best fit with the data will be utilized. Diagnostics of the model agreement with the data was carried out based on the standardized values of the conventional indicators CFI, RMSEA, and SRMR and the suitability of the model was assessed based on selected criteria according to the recommendations of Kline (2015) and Hooper et al. (2007) (CFI>.95; SRMR < .08; RMSEA < .08). Reliability estimation was determined using McDonald’s omega coefficient.

The possibility of predicting general life satisfaction and the frequency of positive and negative experiencing based on gender, age, cohabitation with the father, and the reported father involvement was determined through hierarchical linear regression. The WebPower software (Soper, 2021) was used to calculate the sample size. For the identification of a medium effect (f² =.15) at the statistical power level of .95 when examining four predictors, 129 respondents represent the minimum number. Regression analyses were controlled for gender, age, and cohabitation with the father (Block 1). In the Block 2, the variable father involvement was added to the model. The final models were checked for meeting the conditions for the use of hierarchical regression. The result of the Durbin-Watson test was not statistically significant and had the following values: general life satisfaction = 1.90; positive experiencing = 2.00; negative experiencing = 2.08. The values 1.50–2.50 (Ho, 2013) were considered acceptable. Variance inflation factor values were in all cases lower than 5 (Midi et al., 2010) and tolerance values were higher than .20 (Menard, 2002). The values of the correlation coefficients (Appendix 1, Table A1) confirmed that the relationships between the independent variables were not collinear (<.90).

Results

Pilot study: the factor structure verification of the Slovak version of the reported father involvement scale

To meet the research objectives, the first part of the analyses focused on verifying the factor structure of the Slovak translation of the Reported Father Involvement Scale. Specifically, the validity of the 3-factor theoretical model (Finley & Schwartz, 2004), its 2-factor alternative (Finley et al., 2008) and 1-factor alternative (González-Calderón et al., 2015) was verified. The results in Table 1 reveal that the compared models show almost identical values of fit indices. Due to the difficulties related to the third factor, which we anticipated in the theoretical section and further describe in the discussion section, we consider the 2-factor model more appropriate and examine it further in more detail.

Table 1. Confirmatory and factor analysis of the reported father involvement scale (N = 400).

	X^2/df	X^2	df	CFI	SRMR	RMSEA
3-factor model	4.06	532.93**	131	.997	.043	.088
2-factor model	4.00	604.84**	151	.997	.044	.087
1-factor model	4.02	610.95**	152	.997	.044	.087

X² -chi-square, df-degrees of freedom, CFI, SRMR, RMSEA = goodness of fit tests, **p < .001

In the 2-factor model, high collinearity was identified between both factors of the father involvement, which was controlled by a variance inflation factor (expressive involvement VIF = 7.55, tolerance - .13; instrumental involvement VIF = 7.87, tolerance - .13). According to Midi et al. (2010), an acceptable VIF value is less than 5 and tolerance values should be greater than .20 (Menard, 2002). Likewise, a significant positive correlation was found between both factors (r = .93; p < .001) (Figure 1).

Figure 1. Saturation of factors by individual items.

Circles represent latent factors (F1 – factor of expressive involvement; F2 – factor of instrumental involvement); numbers above items represent standardized estimates; numbers below items represent errors.

In connection with the above mentioned finding of high collinearity, we decided for the next step – verifying the validity of the 1-factor model. The regression weights expressing the relationships between the items and the factor ranged from .63 to .93. The factor was highly saturated with individual items, with item 15 (‘protection’; .95) being the most saturated item. The values of factor saturation by individual items are presented in Figure 2. As in the case of the 2-factor model, the anticipated nonsignificant Chi-square value (x² = 611.120; df = 152; p < .001; chi-square/df = 4.02) was not proven. The justification for accepting this result is already stated above. The values of other indices (CFI = .997; SRMR = .044; RMSEA = .087; 90% CI [.080, .094]) except for RMSEA indicate a reasonably good fit of the model with the data (Browne & Cudeck, 1992). The RMSEA value indicates a mediocre fit of the model with the data (MacCallum et al., 1996).

Figure 2. Saturation of the factor by individual items.

Circles represent latent factors (F1 – father involvement); numbers above items represent standardized estimates; numbers below items represent errors

The 1-factor model showed almost the same values of the individual indices as the 2-factor model, but due to the aforementioned high collinearity of the factors, we decided to use the single-factored Reported Father Involvement Scale in further analyses. For instance, González-Calderón et al. (2015) proceeded similarly in their study (χ2 = 359.00, d.f. = 168; p = .00; χ2/d.f. = 2.13; index values: RMSEA = .048; 90% CI [.041, .055]; GFI = .98; and AGFI = .98).

Main study: the role of perceived father involvement in adolescents experience of subjective well-being

Descriptive and correlation analyses

An overview portraying the results of descriptive analyses that cover monitored variables is provided in Table 2. The skewness of the variables ranges from −0.67 to 0.27, which is adequate for the use of planned statistical analyses.

Table 2. Descriptive characteristics of examined variables.

Variables	M	SD	Skew	Range	McDonald’s ω	AVE	Items
Subjective Well-Being
GLS	4.33	0.92	−0.67 (0.12)	1–6	.82	.53	7
PE	4.08	0.89	−0.37 (0.12)		.83	.59	4
NE	3.05	0.95	0.27 (0.12)		.90	.52	6
Father Involvement
INV	3.36	1.11	−0.63 (0.12)	1–5	.97	.85	19

GLS – General life satisfaction; PE – Positive experiencing; NE – Negative experiencing; INV – father involvement; M – mean; SD – standard deviation; Skew – skewness; Range – the range of the rating scale; McDonald’s ω – reliability estimate; AVE – average variance extracted

Moreover, Table 2 presents the values of estimated reliability (McDonald’s ω) and convergent validity (AVE – average variance extracted, AVE > .50) as basic characteristics for assessing the quality of measurement within structural models. Single-item constructs were omitted from this part of the analysis. The results indicate that the requirements have been met.

In Appendix 1 we present the correlations between the potential predictor (the father involvement) and individual criteria (indicators of subjective well-being). Statistically significant relationships were demonstrated in all cases. The father involvement was significantly positively correlated with general life satisfaction (r = .45; p < .001) and positive experiencing (r = .41; p < .001). It was significantly negatively correlated with negative experiencing (r = −.40; p < .001).

Results of regression analyses

Regression model for predicting general life satisfaction in adolescents

An overview of the results of the hierarchical linear regression analysis for the prediction of general life satisfaction as one of the indicators of subjective well-being is presented in Table 3. The controlled demographic variables included in the Block 1 explained 10% of the variance of general life satisfaction (F = 14.60; p < .001), while gender (β = .14; p < .05) and cohabitation with the father (β = .26; p < .001) proved to be significant predictors. The predictive power of the model increased after including the father involvement in the Block 2 by 11.90% (to 21.90%, F = 27.6; p < .001) of the explained criterion variance, and the father involvement became the only significant predictor (β = .41; p < .001) of general life satisfaction in this step.

Table 3. Results of regression analyses for experiencing general life satisfaction.

Predictor	Block 1			Block 2
	b (CI)	SE (b)	β	b (CI)	SE (b)	β
Gender	1.97 (0.62; 3.33)	.69	.14*	1.06 (−0.20; 2.37)	.65	.08
Age	−0.45 (−1.02; 0.11)	.28	−.08	−0.17 (−0.71; 0.37)	.27	−.03
Cohabitation with father	3.82 (2.45; 5.19)	.70	.26**	0.97 (−0.50; 2.44)	.75	.06
Father involvement				0.13 (0.09; 0.16)	.02	.41**
	R^2 .100**			ΔR^2 .119**
	F 14.3**			ΔF 59.8**

*p < .05 ** p < .001; SE(b) – unstandardized coefficient; β – standardized coefficient; R² – determination index; F – statistical significance of the model; ΔR² – change in determination index; ΔF – statistical significance of the change.

Regression model for predicting general positive experiencing in adolescents

Table 4 shows the results of the hierarchical regression analysis for the prediction of positive experiencing as another indicator of subjective well-being. The controlled variables included in the Block 1 explained only 5% of the variance of positive experiencing (F = 6.96; p < .001), with gender (β = .15; p < .001) and cohabitation with the father (β = .13; p < .05) being significant predictors. The father involvement added in the Block 2 explained 13.3% more of the variance. Thus, the overall model explained 18.3% (F = 22.00; p < .001) of the variance in positive experiencing, while only gender (β = .10; p < .05) and the father involvement significantly increased positive experiencing of adolescents (β = .43; p < .001).

Table 4. Results of regression analyses for positive experiencing.

Predictor	Block 1			Block 2
	b (CI)	SE (b)	β	b (CI)	SE (b)	β
Gender	1.30 (0.54; 2.06)	.39	.15**	0.78 (0.06; 1.50)	.37	.10*
Age	−0.19 (−0.51; 0.13)	.16	−.06	0.02 (−0.28; 0.32)	.15	.01
Cohabitation with father	1.05 (0.28; 1.82)	.39	.13*	−0.61 (−1.43; 0.09)	.42	−.08
Father involvement				0.02 (−0.07; 0.10)	.01	.43**
	R^2 .050**			ΔR^2,133**
	F 6.96**			ΔF 63.80**

*p <,05 ** p <,001; SE(b) – unstandardized coefficient; β – standardized coefficient; R² – determination index; F – statistical significance of the model; ΔR² – change in determination index; ΔF – statistical significance of the change.

Regression model for predicting general positive experiencing in adolescents

The results of the hierarchical regression analysis are presented in Table 5. Demographic variables in the Block 1 explained 12,4% (F = 19.00; p < .001) of the variance of the variable, while gender (β = −.25; p < .001), age (β = .15; p < .001), and cohabitation with the father (β = −.19; p < .001) proved to be significant predictors. After adding the father involvement factor in the Block 2, the predictive power of the model increased by 8.60%. Gender (β = −.19; p < .001) as well as age (β = .11; p < .05) were again significant predictors (girls inclined to negative experiencing more than boys, the frequency of experiencing negative emotions increased with age) and at the same time the father involvement significantly reduced the frequency of negative experiencing (β = −.35; p < .001) in respondents. The overall model explained 21.30% of the variance of negative experiencing (F = 26.60; p < .001).

Table 5. Results of regression analyses for negative experiencing.

Predictor	Block 1			Block 2
	b (CI)	SE (b)	β	b (CI)	SE (b)	β
Gender	−3.12 (−4.30; −1.94)	.60	−.25**	−2.44 (−3.59; −1.30)	.58	−.19**
Age	0.86 (0.36; 1.35)	.25	.15**	0.58 (0.10; 1.06)	.24	.11*
Cohabitation with father	−2.49 (−3.68; −1.29)	.61	−.19**	−0.33 (−1.63; 0.98)	.66	−.03
Father involvement				−0.09 (−0.12; −0.07)	.01	−.35**
	R^2 .126**			ΔR^2 .086**
	F 19.00**			ΔF 43.20**

*p <,05 ** p <,001; SE(b) – unstandardized coefficient; β – standardized coefficient; R² – determination index; F – statistical significance of the model; ΔR² – change in determination index; ΔF – statistical significance of the change.

Discussion and conclusion

To fulfill the main objective of the research, which was to verify the relationship between the reported father involvement and the subjective well-being of adolescents, the first step was to pilot test the factor structure and estimate the reliability for the Slovak translation of the Reported Father Involvement Scale (Finley & Schwartz, 2004) (Pilot study). The original scale is intended for an adult population to retrospectively assess the father involvement. In the presented research, it was used for the adolescent population. For this age category (15–19 years), it had been successfully used, for example, in the validation study of the Spanish version of the questionnaire (González-Calderón et al., 2015). As we indicated in the introduction of the study, the authors of the scale originally proposed a 3-factor structure based on a theoretical model (Finley & Schwartz, 2004), where involvement as a construct consisted of three related factors, namely expressive and instrumental involvement of the father and mentoring/advising. However, the third factor consisted mostly of items that overlap between the factors and complicate the interpretation, despite the model (as well as the 2-factor model) showing a good fit with the data. For this reason, a simpler 2-factor model was analysed in a later study by the authors of the scale (Finley et al., 2008), which no longer included the mentoring subscale. The authors proceeded similarly when validating the Spanish version of the scale (González-Calderón et al., 2015) with the justification that the mentoring/advising subscale was a conceptual overlap between the instrumental and expressive involvement of the father. Based on this knowledge, we initially focused mainly on the verification of the 2-factor model, which was compatible with the original theoretical construct and was compared with the original 3-factor model for completeness. The results of confirmatory factor analysis executed on our dataset indicate acceptable support (MacCallum et al., 1996) for both models. However, given the conceptual and interpretive purity and critical discussion that was conducted in the case of the third factor (Finley et al., 2008; González-Calderón et al., 2015), we found the 2-factor model more appropriate in this comparison. Considering the finding of significant common variance in both factors of the father involvement (Figure 1), we decided to verify the 1-factor model as well. As the results revealed, this model showed the same index values as the 2-factor model.

In view of the aforementioned high collinearity of the factors, we decided to use the single factor scale in further analyses. Our procedure is in accordance with, for instance, the procedure of González-Calderón et al. (2015) that was used in a study aimed at the adaptation of the Spanish version of the scale, in which the 1-factor model prevailed. In a similar way (i.e. as a single factor), the scale was also used by the authors Alifa and Handayani (2021) in research examining the relationship with the subjective well-being of adolescents. Likewise, in the most recent validation study of the Portuguese version of the scale (Cracco et al., 2022), the authors recommend a 1-factor model based on the results of exploratory and subsequent confirmatory factor analysis. We consider the question of verifying the factor structure of the scale to be still open, especially in view of the threshold values of one of the indices and the ambiguity of the results of the confirmatory analyses when comparing the three models. The possibility of verifying the existence of a latent factor and specific variance for individual involvement factors (Bi-factor model SEM) using structural modelling seems also perspective.

Regarding reliability estimation, acceptable values were found. The coefficient of internal consistency exceeded the generally accepted value of > .70 in our research, as was the case in several other studies (Alifa & Handayani, 2021; González-Calderón et al., 2015).

In connection with the main objective of the study, based on the results of the correlation analyses, we can state that the father involvement as a predictor is fully related to the indicators of subjective well-being (Main study). Predicted significant positive relationships were confirmed between the father involvement and general life satisfaction – which is in line with, for instance, the results of Ozdemir (2012) – but also between the father involvement and the positive experiencing of adolescents, as stated in the study of Allgood and Beckert (2012). The expected significant negative relationship was demonstrated between the father involvement and negative experiencing, similar to research by Ozdemir (2012) or Yap and Baharudin (2016). In addition, we were also interested in the relationships between cohabitation with the father and his involvement. Here, positive relationships were confirmed, indicating that cohabitation with the father in a joint household may be related to the perception of the father as more involved by adolescents. Similar results are reported in the study of, for instance, Flouri (2008), and Little et al. (2019) also point out that, in addition to cohabitation with the father in a joint household, the intensity and way the father is involved in the adolescent’s life also plays a role.

Based on regression analyses, the effect of the father involvement was confirmed as a significant predictor of all three selected indicators of subjective well-being. The obtained results indicate that the more adolescents perceive their father as involved in their upbringing, the higher their positive experiencing rate is, the lower their negative experiencing rate is, and they show a higher level of general life satisfaction. We find support for these findings, for example, in the explanation of Yap and Baharudin (2016), who pointed out that it is the emotionally warm and supportive style of fatherhood associated with spending time together with adolescents that is related to an increase in subjective well-being of adolescents. The positive effect of the father involvement on the overall experiencing of subjective well-being in adolescents has been confirmed in several studies with different cultural contexts (e.g. Alifa & Handayani, 2021; De Goede et al., 2009; Flouri, 2006; Piko & Hamvai, 2010; Pitsoane & Gasa, 2018; Sağkal et al., 2018; Yap & Baharudin, 2016).

It is vital to note that the degree of variability explained by the selected variables was not high (18,3% − 21,3%) in any of the cases. However, subjective well-being is a multi-causally conditioned phenomenon, which is influenced by various psychological and non-psychological factors, therefore we did not expect a high percentage explanation of the variance. The same is the case in other research studies with the same focus (e.g. Allgood & Beckert, 2012).

Among the controlled variables, only two variables (gender and age of respondents) became significant predictors in the final models, specifically gender in the case of positive experiencing and both gender and age in the case of negative experiencing. Girls showed a higher tendency to experience both positive and negative emotions compared to boys, similar to the findings of, for instance, Garcia (2011) or Yeo et al. (2007). Esteban-Gonzalo et al. (2020) emphasize that girls generally show a higher level of emotionality during adolescence, whether it is experiencing negative or positive emotions. In this case, age appeared as a significant positive predictor. The probability of the occurrence of negative emotions increased with the age of the respondents. A possible explanation is offered by research studies (e.g. Kohútová et al., 2021; Salmela-Aro & Tuominen-Soini, 2010), which point out that the period of entering adulthood is a challenging period associated with negative experiencing.

Benefits, limitations, and perspectives for further research

We consider the thematic focus of the conducted research on a relatively new, under-researched construct of the father involvement to be beneficial. Furthermore, we perceive the pilot verification of the factor structure and reliability estimation of the Slovak version of the scale on the adolescent population as a favourable step. Nevertheless, for its further use, in addition to the repeated verification of the factor structure via the possibilities indicated in the discussion section, a more detailed examination of the validity is necessary since our research covers the issue only partially (by examining construct validity and AVE as one of the parameters of convergent validity). Another limitation regarding the validity of the findings can be considered the cross-sectional nature of the research and the fact that the relationships found cannot be interpreted causally. In addition, since it was a correlational non-experimental study, it cannot be excluded that the tested relationships are influenced by other variables that were not included in the regression models. Another limitation of the study is the fact that the data collection took place unintentionally at the onset of the 2^nd wave of the pandemic, so there was again a significant restriction of going out and online teaching. Long-term intensive contact of family members in a joint household or, on the contrary, limited contact with the father with whom the adolescents did not live in a joint household could in different ways affect the sensitivity of perception towards the attitude and degree of involvement of the father and parents in general (e.g. Bačíková-Slešková et al., 2021; Donker et al., 2021). We could not compare the data obtained in these specific conditions with the data from the same persons gathered before the pandemic. The online collection itself also brings several limitations (impossibility of direct inquiry, additional explanation, if necessary, etc.).

In conclusion, it can be concluded that the results of our study confirm the validity of examining the father’s involvement in the context of the subjective well-being of adolescents, while practical point of view we perceive as a perspective a more detailed examination of what specifically adolescents understand by the father involvement in their lives, in accordance with the results of the research of Šeboková (2018). At the same time, the role of other factors could be considered in the future, such as parenting style of both parents (Temmen & Crockett, 2021), coparenting (Teubert & Pinquart, 2011), self-efficacy, and emotional control of adolescents (Yap & Baharudin, 2016). These could mediate or moderate the relationship between the perception of the father involvement and the subjective well-being of adolescents.

Ethics statement

Ethical approval for the study was firstly obtained from the Ethical Committee of the Faculty of Arts of the University of Pavol Jozef Šafarik in Košice, Slovakia. Active consent was asked from students.

Acknowledgments

We thank all the schools and students for participating in this study.

This article was supported by VEGA grants: VEGA 1/0719/20: The role of the father in the social, cognitive and emotional development of the child in the current society; Specifics of the transition to adulthood in the context of goals and selected personality variables and their impact on the subjective well-being of young adults.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data, analysis script, supplementary tables and descriptions are available at the OSF (in the text of the manuscript we provide a link where it is not possible to identify the authors; https://osf.io/xctsw/?view_only=96bb836679fa408fbee0da4bb357553b).

Additional information

Funding

This work was supported by the Vedecká Grantová Agentúra MŠVVaŠ SR a SAV [1/0719/20].

Notes on contributors

Martina Semešiová

Martina Semešiová is a PhD. student of Psychology in the Department of Psychology, Faculty of Arts at the Pavol Jozef Šafárik University in Košice. With a focus on developmental psychology.

Beáta Ráczová

Beáta Ráczová is anassociate professor and Vice-Dean for Education of Faculty of Arts,University of Pavol Jozef Safarik in Košice, Slovakia. PreviouslyHead of Department of Psychology, Faculty of Arts, UPJŠ in Košice.Her research interests include the topic of perception and experienceof developmental transitions and the implementation ofself-regulation within these transitions.

Peter Babinčák

Peter Babinčák is aprofessor at the Institute of Psychology, Faculty of Arts, Universityof Prešov. His research interests include the subjective quality oflife from both an individual and family perspective, as well as theissue of fatherhood and the role of fathers in a child’s well-being.

Appendix Appendix 1.

Correlation matrix of examined variables

Table A1. Correlation matrix of examined variables.

Variables	Gender	Age	Cohabitation with father	INV	GLS	PE	NE
Gender	—
Age	rbbis=.04	—
Cohabitation with father	rbbis=.08	rbbis=.01	—
INV	rbbis=.19**	r=−.14*	rbbis =.50**	—
GLS	rbbis=.15*	r=−.07	rbbis =.27**	r=.45**
PE	rbbis=.17**	r=.04	rbbis =.14*	r=.41**	r=.64**	—
NE	rbbis=-.25**	r = 14*	rbbis=-.21*	r=−.40**	r=−.61**	r=−.45**	—

*p < .05 ** p < .001; INV – Father involvement; GLS – General life satisfaction; PE – Positive experiencing; NE – negative experiencing; r_bbis – point-biserial correlation coefficient values. r – Pearson correlation coefficient values.