Development and First Validation of the Perceived Young Adult Lockdown Parental Relationship Scale (PYALPRS): An Italian Case Study

Abstract

The COVID-19 pandemic and the confinement experience have significantly affected the relationship between young adults and their parents. The present study focuses on the design and validation of the Perceived Young Adult Lockdown Parental Relationship Scale (PYALPRS), a measure assessing two dimensions of the child–parent relationship associated with the COVID-19 lockdown period: Oppression-Conflict and Closeness-Support. After a phase of construct definition and item design and purification, 100 Italian young adults (M = 24; SD = 3.9) were recruited to explore the factor structure of the scale. Then, a sample of 259 Italian young adults (aged 18–35; M = 24; SD = 3.8) was used to demonstrate the psychometric validity of the scale. The results of our confirmatory factor analysis, which resulted in high goodness of fit scores, support two identifiable factors reflecting the theory-based constructs of the PYALPRS. Moreover, internal consistency and convergent and divergent validity analyses show that the PYALPRS can be considered a reliable and valid instrument. ANOVA demonstrated that there were significant differences between being a cohabitant or single as well as between different perceptions of the home space during lockdown on the Oppression-Conflict dimension, while a larger home space perception was associated with the Closeness-Support dimension.

1. Introduction

The COVID-19 pandemic has imposed major health, economic and socio-relational upheavals in most parts of the world. In the long term, it will take the form of a traumatic event of a collective nature [1,2,3] with significant consequences, as has happened in the past [4,5]. The negative effects of the pandemic on mental health have been addressed in the literature [2,6,7,8,9], particularly among adolescents and young adults [10,11,12,13,14,15,16,17,18]. The latter were defined early as an at-risk target for several forms of psychological distress, including internalizing and externalizing symptoms [19,20,21,22,23]. The management of developmental tasks specific to this age group [24] has been complicated by the pandemic. This is probably due to specific features of the contemporary world, such as the guarantor crisis, culture of urgency, control, and performance, which have long been reported as potential factors of psychic fragilization [25,26]. In addition, the previous literature highlighted a profound crisis of the paternal function which, alongside a process of horizontalization, makes it increasingly difficult for adults to set boundaries to the new generation they are responsible for [27,28,29,30,31,32,33]. This function crucially modulates the aggressivity and anguish typical of the adolescent developmental process [34,35]. In an era when narcissistic components predominate [36,37], growth paths seem to be lacking the formative experience of recognizing behavioral expectations and limits in all their forms. Therefore, a certain psychic fragility in dealing with setbacks and obstacles might have led young adults to experience the COVID-19 pandemic in a more traumatic way, as it represented a condition of freedom limitation and boundary enforcement par excellence [1,38,39,40,41].

In this context, the COVID-19 pandemic and the major restrictions imposed by the lockdown compromised young adults’ process of separation from their families [42,43]. Returning home was experienced as a dysfunctional regressive movement, leading to a sense of failure marked by the loss of autonomy, control, and a positive representation of the future [44,45,46]. Forced cohabitation in families exacerbated conflict among their members, disrupting preexisting balances and forcing family reorganization [47,48,49,50]. This concept becomes particularly clear when interpreted through the lens of family systems theory [51], which posits that families are units in which each member contributes to maintaining the integrity of the system. Due to COVID-19 pandemic restrictions, families had to rapidly adapt to unprecedented circumstances and face unique stressors—such as the shift to remote learning and employment disruptions—which profoundly affected their daily functioning [52].

Considering this, life course theory [53] helps further explain the pandemic’s effects on family ties through the notion of “linked lives”. This concept refers to the network of social relationships in which each individual is embedded, emphasizing that the effects of historical events, such as the COVID-19 pandemic, must be understood in light of this interconnected network. From this perspective, the stressful changes parents had to face, including disruptions to work and finances, may have undermined their psychological resources, thereby influencing their parenting approaches and the support they could offer their children.

In addition, the application of family stress theory [54] to the pandemic event provides a deeper understanding of the complex interaction between the stressor event, family resources, a family’s perception of the stressor, and the stressor outcomes. According to this approach, when families lack the resources to cope with a stressor event in an adaptive manner, they experience it as a crisis. This, in turn, influences the family’s response to the stressor, including changes in parents’ attitudes and behaviors toward their children. The more recent COVID-19 Family Disruption Model [55] expands on the cascading effect of the pandemic’s outbreak on family functioning, focusing on the buffering role of family processes, such as organization, communication and shared beliefs, on enhancing resilience and buffering the social disruption’s influence on families’ well-being.

In many cases, these processes were hindered by pandemic-related modifications, such as altered cohabitation conditions, which were so overwhelming that young people had to develop new ways to cope with the stressful context. In this vein, a study by Dotson et al. [56] highlighted the affective distancing strategy, enacted by young adults to preserve a mental space which was not easy to find at home. Moreover, several studies emphasize the difficulties young people experienced with their parents during the pandemic. These included a perceived parental inability to understand and contain their boredom and sense of emptiness as well as to communicate regarding pandemic restrictions and the perception that support and closeness from loved ones could not reduce pandemic-related stress and symptoms [57,58,59]. Furthermore, it was found that the relationships between young adults and their parents were marked not only by a strong ambivalence between conflict and support but also by inexperience with the adult world as the main source of support [44,45]. Some studies reported that the lockdown experience exacerbated conflict dynamics and dysfunctional relationship patterns in the home environment by imposing new family and work patterns [60,61]. However, a good parent–child relationship, based on structured mutual communication skills in adolescence and young adulthood [62], has been found to attenuate the occurrence of internalizing and externalizing problems and strengthen parental closeness [63] during lockdown [62,63,64]. According to an intergenerational solidarity model framework [65], the COVID-19 pandemic may have simultaneously fostered and disrupted cohesion between generations due to its complex nature. For example, cohabiting during COVID-19 lockdowns may have allowed young adults and their parents to benefit from mutual social support, but it also required them to consider the costs associated with their proximity, including potential health risks [66].

The integration of the results of various studies about the relationship between parents and children during the pandemic, including our preliminary investigation [45], motivated the creation of the Perceived Young Adult Lockdown Parental Relationship Scale (PYALPRS). This instrument investigated two polarities of the parent–child relationship during the second phase of the pandemic in Italy: Oppression-Conflict and Closeness-Support. In particular, the first dimension explores young people’s perceptions of the degree of conflict and oppression in their relationships with their parents, such as feeling crushed by their worries and judgments. The second dimension explores the perception of parents’ support and closeness, such as their ability to make their children feel supported, able to share activities, and confide in them during the challenging confinement period.

These two dimensions can be explored through various perspectives and theories. Among them, attachment theory focuses on how infant–parent affective bonds influence individual development and functioning over one’s lifespan [67]. Depending on whether children refer to their parents as a source of security, especially in challenging situations, attachment relationships are categorized as either secure or insecure. While people with a secure attachment are more likely to ask for comfort and help from their parents in stressful conditions, insecurely attached individuals find it difficult to deal with negative emotions by referring to their parents and are prone to engaging in dysfunctional conflict resolution strategies [68,69,70].

These dynamics become particularly relevant during potentially traumatic events like the COVID-19 pandemic, where increased stress and uncertainty enhance the importance of attachment bonds and underscore the role of parent–child relationships.

Several pandemic studies have shown how attachment security was a protective factor against psychopathology, highlighting the relevance of close and supportive relationships with attachment figures—rather than insecure ones—in facing the experience of social isolation [71,72,73].

Aligned with attachment theory, the integrated resilience framework proposed by Masten and Palmer [74] emphasizes the critical role of parents in nurturing and protecting children during early development, as this process cultivates adaptive systems which individuals will rely on throughout life, fostering resilience at both individual and societal levels.

The theoretical background and studies presented highlight the relevance of deepening the understanding of family dynamics in times of stress and their broader implications on children’s developmental trajectories and well-being.

The instrument presented here can prove useful in investigating parent–child relationships in general and, more specifically, in other emergencies which might force cohabitation, such as health emergencies or natural disasters like the recent flood in Emilia Romagna.

Aim of This Study

The present research is part of a larger project which investigated the pandemic experience in Italian young adults from the first lockdown in March 2020 to June 2021 through a bottom-up survey approach within a research intervention setting.

The scale presented in this study is part of a broader battery of instruments, the Young Adult Pandemic Experience Questionnaire (YAPEQ), which aims to explore several themes and constructs linked to different phases of the pandemic, including significant concerns, the prevalence of positive or negative feelings, perceived personal restrictions, infodemics, and trust in institutions.

This work describes the development and validation process of the Perceived Lockdown Young Adult Parental Relationship Scale (PYALPRS), an instrument for assessing the perceived quality of the relationship between young Italian adults and their parents with reference to the lockdown period. Specifically, Study I aimed to explore the latent structure of the PYALPRS through exploratory factor analysis (EFA) and assess its preliminary psychometric properties. Subsequently, an independent sample (N = 259) was recruited for Study II to confirm the dimensional structure of the PYALPRS through confirmatory factor analysis (CFA), examine its internal consistency, evaluate the measurement invariance of the instrument with respect to gender, and assess its convergent and discriminant validity.

2. Materials and Methods

2.1. Procedure

Participants for the two studies were recruited in Italy through social media pages between May and June 2021. Convenience and snowball sampling methods were employed based on the following criteria: age between 18 and 35 years old, Italian nationality, and residence in Italy. All data were acquired through self-reporting questionnaires using an Internet-based survey [75]. The participants took part in the study voluntarily, aware of their anonymity, without receiving any incentives. All participants signed a consent form on the first page of the survey which included information about the aims and procedures of the study, confidentiality, and the anonymity of the answers. Those who did not meet the criteria and did not give consent were excluded. The survey took approximately 20 min to complete.

Following Spector [76], the PYALPRS was developed by following these steps: (1) construct definition; (2) item and scale design; (3) pilot testing the questionnaire; (4) questionnaire administration and measure purification; and (5) construct validity verification. Furthermore, for the design of the instrument, deductive and inductive approaches were integrated according to the recommendations of Boateng et al. [77].

The literature review of young adults’ relationships with their parents during the COVID-19 confinement highlights two main emerging polarities: Oppression-Conflict and Closeness-Support [50,78]. A qualitative study was conducted to define the construct before generating the items of the instrument. In this study, 23 testimonial narratives were collected and subsequently analyzed through interpretative phenomenological analysis (IPA) [79]. The narratives show ambivalent relationships with the parents during the confinement period. On the one hand, there are difficulties in communicating, sharing, and managing spaces and perceived parental overcontrol, especially for those children who had previously left their family homes and found themselves forced to return there due to the pandemic. On the other hand, there is a sense of protection coming from parental support and a newfound variety of activities to spend time together, which have enhanced mutual understanding and support.

The findings of these qualitative analyses, which were further examined by Regnoli et al. [45], corroborated the presence of the two main polarities (Oppression-Conflict and Closeness-Support), which were used to guide the creation and selection of the PYALPRS items.

Once a definition was outlined, we developed an initial pool of 16 items (8 items for each dimension), consistent with the conceptual definition of young adults’ relationships with their parents during the COVID-19 confinement. All items were judged by 3 independent experts who were provided with the definition of the proposed construct and asked to identify any ambiguities in the item wording or any inconsistencies between the item and the dimension to be captured. No expert found any ambiguity in item wording, while 2 experts found that 6 items were incompatible with the dimensions to be measured. Based on this, experts were then asked to more strictly express their agreement about the consistency between each of the remaining 10 items and the dimension to which each was assigned according to a 7-point Likert-type scale, ranging from 1 (“strongly disagree”) to 7 (“strongly agree”). Interrater reliability, determined by using the intraclass correlation coefficient, was good (ICC = 0.85) (<0.5 = poor; 0.5 < moderate < 0.75; 0.75 < good < 0.9; >0.9 = excellent [80]). Thus, the obtained version of the PYALPRS consisted of 10 items reliably categorizable into one of the two dimensions.

Considering 10 participants per item [81], we administered this 10 item scale to the first sample to explore the latent structure of the PYALPRS with the aim of selecting the items by deleting those with redundancy and comprehension problems.

Then, the final pool selected was administered to the second sample to confirm the psychometric structure and validity of the PRYALPS. A sociodemographic section was added to the scale to explore information regarding participant gender and age, marital and employment status, region of residence, type of cohabitation during confinement, and space perception during confinement.

2.2. Sample Size Determination

The a priori sample size planning followed the criterion of 10 subjects per item for both Study I and Study II, in accordance with the recommendations of Mundfrom et al. [81] and Terwee et al. [82].

2.3. Participants

For Study I, a sample of 100 young adults (M = 24.0; SD = 3.92) completed an initial pool of items to evaluate the latent structure of the scale. Then, for Study II, a second sample was employed to validate the final version of the PYALPRS. This sample consisted of 259 young adults aged between 18 and 30 years (M = 24.3; SD = 3.82). Most participants of the final sample were female (n = 198; 76.4%), university students (n = 200; 77.2%), undergraduate students in humanities (n = 193; 74.5%), and in a relationship (n = 123; 47.5%). Most of participants declared cohabiting with their families during the confinement (n = 205; 79.2%). Furthermore, they were predominantly from southern Italy (n = 204; 79.0%) and shared living spaces with their families during confinement (n = 205; 79.2%).

2.4. Measures

In addition to the PYALPRS, the following instruments were used to examine the data.

A sociodemographic questionnaire was constructed to collect information regarding participant gender and age, region of residence, place of residence (town or country), marital status, level of education, employment status, shared living spaces with family during confinement, and type of cohabitation.

The Lockdown Young Adult Concerns Scale (LYACS) [83] is an instrument assessing concerns during the pandemic lockdown. The LYACS is a 5-point Likert-type scale ranging from 1 (“not at all”) to 5 (“very much”) which measures two dimensions: Loss of Life Control (CLLC) and Infection/Contagion (CIF). In this study, an overall score was used. Regnoli et al. [83] reported good internal consistency for the instrument. In the current study, Cronbach’s α and McDonald’s ω were 0.81 and 0.79 for the overall scale, respectively.

The Satisfaction with Life Scale (SWLS) [84], with Italian adaptation and validation by Di Fabio and Busoni [85], is an instrument developed to assess global life satisfaction. It is a 7-point Likert-type scale composed of 5 items ranging from 1 (“strongly disagree”) to 7 (“strongly agree”). Di Fabio and Busoni [85] reported good internal consistency. In the current study, Cronbach’s α and McDonald’s ω were 0.90 and 0.90, respectively.

The Life Orientation Test-Revised (LOT-R) [86], with Italian adaptation and validation by Anolli [87], is an instrument which measures dispositional optimism. It is a 5 point Likert-type scale composed of 10 items ranging from 1 (“strongly disagree”) to 5 (“strongly agree”). Anolli [87] reported good psychometric proprieties for the instrument. In the current study, Cronbach’s α and McDonald’s ω were 0.82 and 0.82, respectively.

2.5. Statistical Analyses

In Study I, exploratory factor analysis (EFA) with principal axis factoring and direct oblimin rotation was used to identify the factor structure of the scale. Eigenvalues >1.0 [88], communality ≥0.30 for each item, and factor loading >0.35 for each item associated with extracted factors [89] were selected as the criteria. The measure of sampling adequacy (MSA ≥ 0.50) was also evaluated to explore the adequacy of the pool items in measuring the specific domain [90]. The Latent Hancock Index (H Index ≥ 0.80) was preliminary calculated to test how well the items of the tool reflected the latent variable with a high probability of stability in several studies. Complementary indices of the Factor Determinacy Index (FDI) (≥0.90), sensibility ratio (SR) (≥2.0), and expected percentage of true differences (EPTD) (≥90.0%) were calculated to test the quality and effectiveness of the factor solution.

In Study II, descriptive analyses were performed for the scale items, including the mean, standard deviation, kurtosis, skewness, and item–total correlation. Descriptive statistics were also calculated for the two subscales of the PYALPRS. Then, to confirm the factorial structure, confirmatory factor analysis (CFA) was conducted using the robust maximum likelihood estimation method (MLM) and goemin rotated solution. The MLM is a robust estimator which provides robust standard errors, and it is referred to as the Satorra–Bentler chi-squared test (SBχ²) to assess model fit. Following the recommendations of Hu and Bentler [91] and McDonald and Ho [92], the indices used to evaluate the goodness of fit of the extracted model were as follows: SBχ²/df (in a range from 2 to 5); the goodness of fit index (GFI; ≥0.90); root mean square error of approximation (RMSEA; ≤0.08) [93]; standardized root mean square residual (SRMR; ≤0.09) [94]; comparative fit index (CFI; ≥0.90); incremental fit index (IFI; ≥0.90); and Tucker–Lewis Index (TLI; ≥0.90).

Invariance of measurement (MI) was assessed to determine whether the factorial structure of the PYALPRS was consistent across females and males. Three nested models were sequentially tested, imposing equality constraints on parameters across the two groups. First, the factor structure was set to be identical across the groups (Model 1: configural invariance). Second, the factor loadings were constrained to be equal across the groups (Model 2: metric invariance). Finally, both the factor loadings and intercepts were constrained to be equal in the two groups (Model 3: scalar invariance). The goodness of fit for each model was assessed using the indices and thresholds outlined earlier, including the SBχ²/df, CFI, TLI, RMESA, and SRMR. Given that the χ² statistic may be more sensitive to the sample size of the comparison groups rather than a lack of invariance, |ΔCFI| has been suggested as the most reliable method for evaluating measurement invariance [95]. Therefore, |∆CFI| (≤0.01) and |∆RMSEA| (≤0.015) were chosen as criteria for assessing gender measurement invariance.

Cronbach’s α [96] and McDonald’s ω [97] were used to test the internal consistency of the PYALPRS while considering acceptable values ≥0.70 [98].

Convergent validity was evaluated by calculating the standardized factor loading (SFL) (≥0.50), composite reliability (CR) (≥0.70), and average variance extracted (AVE) (≥0.50) [99]. To further investigate the convergent and predictive validity of the instrument, Pearson’s correlation analyses were carried out (p < 0.01) with other validated instruments.

Discriminant validity was evaluated through two different methods: Fornell and Larker’s criterion and the heterotrait-monotrait (HTMT) ratio. To assess the ability of a latent variable to differentiate itself from others within the model, the AVE should exceed the shared variance between constructs as measured by the squared factor correlation, according to Fornell and Larcker [100]. Furthermore, the HTMT ratio based on the multitrait-multimethod matrix (MTMM) was considered. An HTMT ratio of less than 0.85 was considered an indicator of good discriminant validity [101].

To explore the possible effects of some sociodemographic variables on the two subscales of the PYALPRS (p value < 0.05), two-way analyses of variance (ANOVAs) were performed. The following variables were considered: gender, marital status, employment status, space perception during confinement, and type of cohabitation. Effect sizes were evaluated through the eta-squared value (η²: small [0.01, 0.059] ≥ 0.01; large [0.059, 0.138] [102]).

Survey data were checked and verified by project staff for accuracy and analyzed using SPSS 27.0 [103], the Factor Analysis Program (available at https://psico.fcep.urv.cat/utilitats/factor/Download.html; accessed on 12 December 2024) and RStudio with the lavaan [104] and semtools [105] packages.

3. Results

3.1. Exploratory Factor Analysis

The results of the exploratory factor analysis indicate that the data were adequate for factor analysis. Indeed, the KMO value was 0.81 (95% CI: 0.78, 0.83), and the result of Bartlett’s test of sphericity was significant [χ² (df = 45) = 437.45 (p < 0.001)]. Inspection of the scree plot and the eigenvalues suggested a two-factor solution.

The initial eigenvalue of the first factor (F1, named Oppression-Conflict [O/C]) was 3.84, while the initial eigenvalue of the second factor (F2, named Closeness-Support [C/S]) was 2.41. The total percentage of variance explained by the two factors was 62.5% (F1 = 38.4%; F2 = 24.1%). All scale items met the selected extraction criteria and were divided as follows. Items 3, 5, 7, and 9 were associated with F1, while items 1, 2, 4, 6, and 10 were associated with F2. The criteria selected for the factor’s extraction suggested considering a two-factor model. The MSA, communalities, and factor loadings for each item are reported in

Table 1. PYALPRS EFA of original 10 items (N = 100).

Items	MSA	h2	F1	F2
3. I felt judged and observed more than I usually am (O/C)	0.83	0.66	0.80	0.05
5. I felt like they/he/she did not understand my struggles (O/C)	0.83	0.77	0.83	0.19
7. I felt overwhelmed by their/his/her worries, anxiety, and fears (O/C)	0.83	0.66	0.81	0.01
8. After an initial phase of conflict, we made compromises and found new common ground (O/C)	0.71	0.35	−0.59	0.21
9. I felt like they/he/she were/was hindering my independence (O/C)	0.86	0.67	0.76	0.21
1. I confided in them/him/her more often than usual (C/S)	0.82	0.73	0.13	0.83
2. I have supported, protected, and reassured them/him/her (C/S)	0.73	0.56	−0.02	0.75
4. I was able to get to know them better (C/S)	0.74	0.67	−0.36	0.79
6. I felt like they/he/she were/was supportive, protective, and reassuring (C/S)	0.81	0.71	0.21	0.79
10. It was good to finally spend some time together and enjoy shared interests and/or activities (C/S)	0.79	0.46	0.13	0.65

Note: Convergence for rotation performed in 8 iterations. MSA = measure of sampling adequacy; h² = communalities; F1 = Oppression-Conflict dimension; F2 = Closeness-Support dimension.

.

For the theoretical consistency of the Oppression-Conflict dimension, we considered it appropriate to eliminate item 8, as it was linked to this factor with a negative sign and with communality just above the cut-off considered. Similarly, since item 4 showed a slight cross-loading on both factors, it was eliminated.

The final PYALPRS administered to the second sample comprised eight items: four items for the Oppression-Conflict factor and four items for the Closeness-Support (see Table 1). This factorial structure demonstrated good preliminary psychometric properties; the latent H indices were 0.88 (95% CI: 0.85, 0.90) for F1 (O/C) and 0.82 (95% CI: 0.75, 0.85) for F2 (C/S), the FDI indices were 0.94 and 0.91 the SRs were 2.73 and 2.15, and EPTDs were 91.7% and 89.9%, respectively.

3.2. Confirmatory Factor Analysis

The means, standard deviations, kurtosis, skewness, and item–total correlation for the items and dimensions of the PYALPRS are shown in

Table 2. Descriptive statistics and confirmatory factor analysis (N = 259).

	Descriptive Analysis					CFA
	M	SD	Sk.	K.	r Item–Total	λ	R2
Item 1	2.4	1.1	0.49	−0.14	0.45	0.672	0.452
Item 2	2.9	1.0	0.13	−0.38	0.32	0.609	0.371
Item 3	3.9	1.2	−0.92	−0.16	0.58	0.808	0.653
Item 5	3.7	1.2	−0.64	−0.73	0.66	0.835	0.697
Item 6	3.0	1.1	0.09	−0.70	0.62	0.846	0.716
Item 7	3.7	1.3	−0.66	−0.67	0.50	0.715	0.511
Item 9	4.0	1.2	−0.97	−0.19	0.62	0.796	0.633
Item 10	3.0	1.8	0.03	−0.71	0.38	0.632	0.340
O/C	3.84	1.04	−0.76	−0.32	-	0.918	-
C/S	2.60	0.73	0.23	−0.43	-	0.498	-

Note: M = mean; SD = standard deviation; Sk. = skewness; K. = kurtosis; CFA = confirmatory factor analysis. In CFA columns, absolute values of standardized factor loading (|λ|) are reported. λ = factor loading onto the specific factor. All λ values are statistically significant with p < 0.001. R² = variance explained; O/C = Oppression-Conflict dimension; C/S = Closeness-Support dimension.

. The first-order model with two interrelated dimensions showed a good fit for the data (Figure 1). Although the chi-squared statistic was found to be statistically significant [SBχ² (19) = 52.691; p < 0.001], the other goodness-of-fit indices indicated a good model fit (SBχ²/df = 2.77; GFI = 0.951; CFI = 0.953; TLI = 0.931; IFI = 0.961; RMSEA = 0.083, 90% CI [0.058–0.010]; and SRMR = 0.061).

All indices used to determine the goodness of fit of the model were adequate.

3.3. Measurement Invariance for Gender

First, configural invariance was tested. The model showed acceptable goodness-of-fit indices, suggesting that the factor structure was equal between males and females (SBχ² (38) = 92.694; p < 0.001; SBχ²/df = 2.44; CFI = 0.932; TLI = 0.900; RMSEA = 0.082, 90% CI [0.074–0.090]; SRMR = 0.069). Then, metric invariance was implemented. The model showed a good fit for the data (SBχ² (44) = 101.969; p < 0.001; SBχ²/df = 2.32; CFI = 0.927; TLI = 0.907; RMSEA = 0.087, 90% CI [0.076–0.101]; SRMR = 0.078). A non-statistically significant decrease in |ΔCFI| (0.005) and |ΔRMSEA| (0.005) was found, indicating that the items were similarly associated with the latent factor regardless of gender. Then, scalar invariance was tested. The model showed a good fit (SBχ² (50) = 11.650; p < 0.001; SBχ²/df = 2.23; CFI = 0.922; TLI = 0.923; RMSEA = 0.090, 90% CI [0.079–0.103]; SRMR = 0.079). A non-statistically significant decrease in |ΔCFI| (0.005) and |ΔRMSEA| (0.001) was found, indicating that the expected item response at the same absolute level of the trait was consistent across both females and males.

3.4. Internal Consistency and Validity

The reliability of the scale, estimated using Cronbach’s α, was 0.80, while McDonald’s ω was 0.77. In detail, Cronbach’s α and McDonald’s ω for Factor 1 (Oppression-Conflict) were 0.87 and 0.87, and for Factor 2 (Closeness-Support), they were 0.77 and 0.78, respectively.

Concerning convergent validity, the standardized factor loadings of the PYALPRS’s items were all greater than 0.50, with ƛ values between 0.632 and 0.846 (see Table 2). The composite reliability (CR) value of Factor 1 (O/C) was 0.87, while it amounted to 0.79 for Factor 2 (C/S). The average variance extracted (AVE) value of Factor 1 (O/C) was 0.79, while it amounted to 0.69 for Factor 2 (C/S). As reported in

Table 3. AVE values of PYALPRS dimensions.

Variables	AVE	R2
		1	2
1. O/C	0.79	-
2. C/S	0.69	0.16	-

Note: AVE = average variance extracted; R² = squared correlations; O/C = Oppression-Conflict dimension; C/S = Closeness-Support dimension.

, the squared correlation between the O/C and C/S dimensions was lower than their respective AVE values. This finding indicates the absence of multicollinearity issues within the measurement, thereby confirming its internal validity. Moreover, the HTMT value of 0.34, obtained from the analysis of the instrument’s item correlation matrix, indicates good discriminant validity, further confirming that the dimensions were significantly different from each other.

Pearson correlations with measures of the Lockdown Young Adults Concern Scale (LYACS), Satisfaction with Life Scale (SWLS), and dispositional optimism (LOT-R) were conducted to test the convergent validity. As indicated in

Table 4. Pearson correlations for PYALPRS regarding convergent and predictive validity (N = 259).

	1	2	3	4	5
1. O/C	--
2. C/S	0.09	--
3. LYACS	−0.33 **	0.14 *	--
4. SWLS	0.29 **	0.28 **	−0.15 *	--
5. LOT-R	0.20 **	0.33 **	−0.22 **	0.57 **	--

Note: * p < 0.05. ** p < 0.01 (two-tailed). C/S = Closeness-Support; O/C = Oppression-Conflict; LYACS = Lockdown Young Adults Concern Scale; SWLS = Satisfaction with Life Scale; LOT-R = Life Orientation Test-Revised.

, the association between the Oppression-Conflict dimension and LYACS was negative and significant (r = −0.33), while the associations between the SWLS (r = 0.29) and LOT-R (r = 0.20) were positive and significant. The Closeness-Support dimension showed positive and significant associations with the LYACS (r = 0.14), SWLS (r = 0.28), and LOT-R (r = 0.33).

3.5. Group Differences

Concerning the mean scores of the subscales, the young adults show medium-to-high Oppression-Conflict scores and slightly above average Closeness-Support scores. A t-test and ANOVA analyses were carried out to explore the principal effects of gender, marital status, employment status, space perception, and type of cohabitation on the Oppression-Conflict and Closeness-Support dimensions.

There were no significant differences in terms of gender (t₍₂₅₇₎ = 0.54; p = 0.29), employment status (F_(5,258) = 2.803; p = 0.11), or type of cohabitation (F_(3,258) = 2.43; p = 0.67) concerning the Oppression-Conflict dimension.

The findings revealed significant effects from marital status [F_(3,258) = 4.56; p < 0.05; η² = 0.05] on the Oppression-Conflict dimension. Post hoc comparisons (Tukey test) suggest that group differences resulted from cohabiting young adults (M = 4.5; SD = 0.4) rather than individuals (M = 3.7; SD = 1.0) because the first group scored higher mean scores than the second one (p < 0.05).

Furthermore, the results show the significant effect of space perception during confinement for the Oppression-Conflict dimension [F_(4,251) = 4.00; p < 0.01; η² = 0.06]. Post hoc comparisons (Tukey test) suggest that group differences were mostly related to the possibility of using an extremely large space (M = 4.4; SD = 0.9) rather than an extremely insufficient space (M = 3.1; SD = 1.2; p < 0.01) or insufficient space (M = 3.6; SD = 1.1; p < 0.05).

Finally, significant effects were observed for the type of cohabitation during lockdown [F_(4,251) = 4.07; p < 0.05; η² = 0.06] on the Closeness-Support dimension. Post hoc comparisons (Tukey test) suggest that group differences were linked to the possibility of using a large space (M = 3.0; SD = 0.9) rather than more than sufficient space (M = 2.4; SD = 0.7; p < 0.05). No significant differences were found in terms of gender (t₍₂₅₇₎ = 1.18; p = 0.12), employment status (F_(5,259) = 1.19; p = 0.31), marital status (F_(3,258) = 1.0; p = 0.40), or type of cohabitation (F_(3,258) = 0.15; p = 0.93) concerning the Closeness-Support dimension.

4. Discussion

The present study shows the development, validation, and evaluation of the psychometric properties of the Perceived Young Adult Lockdown Parental Relationship Scale (PYALPRS), an eight-item self-reporting instrument which explores two relational polarities connoting the perceived quality of relationships between young Italian adults and their parents with reference to the lockdown period [45,46,50]. The scale investigates the parent–child relationship by considering not only difficulties which are specifically related to the pandemic, such as confinement and social isolation, but also those related to developmental tasks in young adulthood.

The results of the exploratory factor analysis (Study I) revealed a two-dimensional structure of the instrument. Two distinct but correlated factors were identified, named Oppression-Conflict (O/C) and Closeness-Support (C/S). The O/C dimension comprises four items focused on understanding how much young adults felt judged by their parents, poorly understood in the difficulties they encountered, and crushed by their worries, anxieties, and fears as well as how much they perceived them as an obstacle to independence during the lockdown (e.g., “I felt like they/he/she were/was hindering my independence” or “I felt like they/he/she did not understand my struggles”). The C/S dimension consists of four items which investigate how much young adults confided in their parents, perceived them as supportive, and enjoyed the time spent together engaging in shared activities (e.g., “I felt like they/he/she were/was supportive, protective, and reassuring” or “I confided in them/him/her more often than usual”).

The results of Study I highlight the significant contribution of all items in defining the two factors, with factor loadings and communalities above the established criteria [89]. Only items 4 and 8 were excluded as they did not meet the selected criteria.

The two extracted factors demonstrated good preliminary psychometric properties. The two factors showed Latent H-Index, Factor Determinacy Index (FDI), sensibility ratio (SR), and expected percentage of true differences (EPTD) values above the recommended thresholds. These results revealed a robust dimensional structure for the instrument which is able to maintain stability across studies with a high probability.

The factor structure emerging from the EFA was confirmed in the confirmatory factor analysis (CFA) conducted in Study II. The results of this analysis highlighted the good fit of the first-order two correlated factor model.

Regarding gender invariance, three models were tested: configural, metric, and scalar. Given the heterogeneity of the two groups in the sample, alternative fit indices were selected to assess the invariance of the instrument, since the chi-squared statistic is often influenced by the sample size [95]. These preliminary analyses confirm the invariance of the instrument between the two groups, supporting the comparison of means and the use of the scale for potential multigroup SEM analyses.

Further analyses revealed good dimensional and global internal consistency, in accordance with the criteria indicated by Cronbach [96] and McDonald [97]. Based on Fornell and Larcker’s criterion [100], the PYALPRS appears to have good convergent validity, with satisfactory scores for the item factor loadings, composite reliability (CR), and average variance extracted (AVE). In addition, the squared correlation between the O/C and C/S dimensions and the HTMT value indicated good discriminant validity for the tool, confirming that the dimensions under investigation were conceptually distinct.

To further investigate the predictive and convergent validity of the instrument, correlations were performed between the two dimensions of the PYALPRS and life satisfaction (SWLS), concerns related to the COVID-19 lockdown (LYACS), and dispositional optimism (LOT-R). The results revealed a significant negative relationship between the Oppression-Conflict (O/C) dimension and concerns associated with the COVID-19 lockdown. This finding suggests that young adults who perceived their parents as more oppressive and conflictual had fewer concerns related to the COVID-19 pandemic (loss of control over life and concern about contagion). This descriptive finding aligns with several studies which highlighted how greater attention to the evolution of the pandemic, information, and health protection advice, as well as increased parental control, reduced pandemic-related concerns within the family [83,106].

Regarding the Closeness-Support (C/S) dimension, the analyses showed significant positive relationships with life satisfaction and dispositional optimism. Concerning the relationship with life satisfaction, our results are consistent with studies which demonstrated how positive relationships with parents and, more generally, within the family are associated with greater life satisfaction (both general and familial), stronger feelings of closeness to parents, and better psychological well-being [107,108]. Regarding the relationship between the Closeness-Support (C/S) dimension and optimism, the results align with several studies indicating that perceiving parents as close and supportive fosters greater optimism and openness to independence [109]. Moreover, perceiving the parent–child relationship positively was found to increase optimism in young people [110].

Our study found a medium-to-high score for the Oppression-Conflict dimension, in line with other studies showing increased stress and conflict between children and parents in situations of forced cohabitation marked by lack of privacy, domestic overcrowding, and increased intra-family psychological distress, even in the pre-pandemic era [58,61,111].

However, our ANOVA results show that conflicts between parents and children do not seem to be associated with the perception of having insufficient space at home but rather with the perception of a large space. In line with the study by Skinner et al. [78], it could be hypothesized that the “home space” variable may not be determinant in conflicts with parents but rather the presence of preexisting difficulties. According to Skinner et al. [78], the quality of family relationships during the pandemic period is associated with the quality of those relationships in the pre-COVID era [59], in line with theories which consider childhood a critical period in the relational development of the individual, like the attachment theory [112]. Several studies have shown that forced cohabitation exacerbated psychological distress and inter-familiar conflicts in young people with prior emotion regulation problems, which were frequently associated with maladaptive strategies in parent–child conflict resolution [78,113]. We believe that another aspect which needs to be investigated is whether and to what extent a high degree of oppression or conflict may be related to the negative experiences typical of young adults who had to return to a nuclear family, as found in other studies [45,46,83]. In fact, although cohabiting may have encouraged social connections and support, it did not eliminate the challenges associated with it [66]. In this regard, family systems theory [51] helps explain how forced cohabitation may have disrupted the routines of both parents and children, altering established family dynamics and requiring their reorganization.

Furthermore, the increased time parents and children spent together sharing the home environment, along with reduced opportunities for peer socialization, may have intensified parental control [114], contrasting with the need for autonomy and independence typical of adolescence and young adulthood [115,116,117,118].

On the other hand, the literature shows that in young adulthood, there is a progressive decrease in conflicts with parents compared with adolescence, particularly when children have already left the family nest [24,114]. Studies on both Italian and North American young adults found that although lockdown-related forced family cohabitation fostered family conflicts, it also promoted closer family ties and a greater sense of safety [115].

These contradictory findings could be explained by considering that the effects of historical events must be understood in relation to the specific characteristics of the networks in which people are integrated, as framed by the “linked lives” principle [53]. As highlighted by family stress theory, during stressful times, families’ experiences are profoundly shaped by their resources to cope with the stressor event, which are crucial in determining their response [54]. In light of this, when examining the findings of pandemic studies, the disruptive potential of the pandemic must be evaluated alongside the quality of family processes [55], which vary from family to family, to better understand variations in the results. Moreover, given the complex nature of the COVID-19 pandemic, it is reasonable that both mutual support and levels of family conflict and stress may have increased [61,78], as highlighted by Gilligan [66].

The results of the ANOVA also showed a positive and significant association between a perceived large home space and Closeness-Support, a sign that this variable played an important role in fostering positive and supportive relationships between parents and children, in line with studies showing that a larger home space during lockdown reduced stress levels [118] and, conversely, a smaller space was associated with greater conflict [61].

In conclusion, it is evident that during the pandemic, strong parent–child relationships protected children and mothers from developing internalizing and externalizing issues [63], consistent with the contribution of attachment theory [67], in the protective role of supportive parent–child bonds developed in earlier stages.

Strengths, Limitations, and Directions for Future Research

To the best of our knowledge, the PYALPRS is the only measure in the literature to investigate the child–parent relationship during the lockdown period, focusing on children’s perceptions of the quality of relationships with their parents. Specifically, such a scale allowed us to explore two polarities of the young adult–parent relationship: Oppression-Conflict and Closeness-Support. The scale showed satisfactory psychometric properties, complied with the validity criteria, and could offer an extra piece of knowledge about the complex traumatic event that is the COVID-19 pandemic.

Despite this, the present research has its limitations. The first limitation is related to the sample. Indeed, convenience sampling can lead to volunteer bias, which concerns the characteristics of individuals who choose to participate in a study or even self-selection bias resulting from the use of social media pages. Additionally, the study’s mono method design could have inflated the observed associations. Furthermore, homogeneity of the sample may have been compromised since most of the respondents were women, students, and from southern Italy. Finally, it is important to recognize that the findings of emergency-related studies are profoundly influenced by the specific timing of data collection. Considering these limitations, the results of this research cannot be generalized to all Italian young adults.

Overall, the analyses carried out show that the presented instrument had good convergent and divergent validity. However, we believe that future studies can deepen and strengthen the promising results described here through correlational analyses and further scales which consider similar and different constructs. In addition, structural equation modeling (SEM) analyses could be conducted to investigate the relationships between PYALPRS factors and several variables, such as academic performance, relationship satisfaction, and mental health indicators. For instance, it would be useful to explore the mediating role of the type of cohabitation or home space perception on the relationship between the Oppression-Conflict and Closeness-Support dimensions and mental health in different social isolation experiences.

Furthermore, longitudinal studies would offer insights into the temporal stability of the scale, assessing whether it can reflect changes in the relationships between young adults and their parents over time. Researchers could explore whether changes in family dynamics during lockdown or other potentially traumatic experiences persist or return to previous levels, providing a more accurate understanding of their impact on future functioning and psychological well-being.

Given the cross-sectional nature of the present study, future research should also explore the psychometric properties of the scale across different countries and cultural settings, providing valuable insights into the scale’s cross-cultural validity.

We also suggest that future research examine the open questions in more depth, perhaps by involving more homogeneous and balanced samples arranged by gender, employment status, and geographical origin.

5. Conclusions

The present study describes the validation process of the PYALPRS, a valid, robust, and reliable instrument which can explore the perception of the quality of the child–parent relationship during the lockdown by detecting Oppression-Conflict and Closeness-Support levels. The proposed scale could be used to explore the relational dynamics between parents and children at critical moments of developmental transition, supporting the in-depth investigation of this complex bond and its transformations in the context of hypermodernity. We also believe that this scale can be useful in exploring young adults’ perceptions of their relationships with their role models and investigating the child–parent relationship in specific collective events with high traumatic potential, such as future health emergencies or natural disasters [119,120,121].

The PYALPRS could play a key role in detecting at-risk family functioning. Moreover, in emergency situations, such as the COVID-19 pandemic, family relationships—particularly parent–child relationships—may play a significant role in either mitigating or amplifying the impact of traumatic or potentially traumatic events. The availability of a brief yet reliable instrument like the PYALPRS could facilitate the assessment of young adults’ perceptions of their parental relationship quality. This, in turn, could support the development of targeted interventions designed to assist this population [122].