Construct Validity of the Portuguese Version of the Wagnild and Young Resilience Scale

THEORETICAL FOUNDATIONS

Resilience is a construct that describes an individual’s capacity to overcome, with relative success, adverse conditions, or situations that involve risk to their well-being, development, and mental health (Reppold, Mayer, Almeida, & Hutz, 2012). Luthar (2006) shows that resilience is a process in which individuals adapt positively to a context which is considered unfavorable and that on adapting themselves, they show a notable and recognized capacity to overcome adverse conditions that could represent a real and significant threat to their well-being, their developmental process, or their mental health. In this sense, and according to Pesce, Assis, Santos, and Oliveira (2004), resilience is understood to be the set of social and intrapsychic processes that make the development of a healthy life possible, even while living in an unhealthy environment. To understand resilience is to accept it as a dynamic developmental process that involves intrinsic and extrinsic factors which contribute to individual differences (Lemery-Chalfant, 2010). Resilient people show a good capacity to rapidly rebalance themselves physically, psychologically, and in social relationships, at times when there are stressful events (Vara & Sani, 2006).

Resilience represents the art of adapting oneself to adverse situations (biological and sociopsychological conditions), developing capacities connected to internal (intrapsychic) and external (social and emotional environment) resources that enable one to unite an appropriate psychic construction and social integration (Anaut, 2005). In this sense, and according to authors Silva, Elsen, and Lacharité (2003), resilience is seen as an individual competence that is constructed based on interactions between the subject, their family, and their environment; and for others, it is a competence not only of the subject but also of some families and certain communities. Thus, resilience “is a multidimensional construct and has three basic requirements: need for a significant risk/adversity; presence of resources to cope with the effects of adversity; and positive adaptation or the impediment of a negative result” (Santana da Silva, Vieira da Silva, Santana da Silva, & Lodovici, 2015, p. 104). For Masten (2001), the concept of resilience takes into account processes of adaptation and risk factors. As far as processes of adaptation are concerned, some disagreements are evident between researchers on resilience. However, there seems to be a consensus regarding the relationship between resilience and risk factors (Cowan, Cowan, & Schulz, 1996).

For Luthar (2006), resilience is characterized by the existence of “adverse conditions” that represent risk factors (often associated with different phenomena of biopsychosocial imbalance) and by the existence of “positive adaptation patterns” of the individual to the risk factors to which they have been exposed. Therefore, there is a need to consider both these factors when analyzing the construct. Resilience therefore requires flexibility and demands the maintenance of a balance in the life of a subject, considering the way he or she deal with and function in the stressful situations and the traumatic events that occur over his or her lifetime (American Psychological Association, 2016).

This brief review allows us to understand not only the enthusiasm but also some theoretical controversies regarding this process and as well the influence of the social and cultural context (Ungar, 2010; Zolkoski & Bullock, 2012). However, it is clear that resilience allows individuals to overcome adverse contexts maintaining positive outcomes (Masten, 2014). For example, in nursing research, literature points out the role of hope, self-efficacy, coping or cognitive reframing as characteristics and strategies that help professionals to maintain their resilience regarding the workload, multiple roles, challenging conditions, and demands from superiors and patients (Cusack et al., 2016; Hart, Brannan, & de Chesnay, 2014).

In the words of Rutter (1990), resilience is concerned with individual variations in response to risk factors being that protective factors enable modification of responses to the situations in accordance with the risk, reducing both the risk and the negative reactions. Therefore, once we understand resilience to be a product of the dynamic interaction of the individual with his or her environment, we have to consider the importance of risk and protective factors. Risk factors are understood to be negative events in life which, when present in the context, increase the probability of the individual developing physical, psychological, or social problems. Protective factors, on the other hand, translate as influences that modify, alter, or improve individual responses to the risks of maladaptation that the individual might suffer. Once we demonstrate that the different factors operate in a dynamic way, we can see that when they interact they modify the trajectory of the individual and may induce a stressful or protective situation.

In this sense, and in accordance with Simões (2008), resilience emphasizes the protective factors and processes that reverse the effects of the risk factors when the individual is exposed to situations that are adverse to healthy behavior.

We must therefore interpret resilience from a dynamic perspective because it is changeable through time and takes into account the circumstances the individual finds themselves in: It is the product of the balance established between protective factors, risk factors, and personality. Rutter, quoted by Pinheiro (2004), shows that to understand resilience, it is extremely important to know how the protective characteristics have developed and how they have modified the trajectory of the individual because the recognition that the existence of protective factors contributes to resilience encourages the development of preventive strategies, which aim at the growth of protective factors and adaptive levels of functioning.

According to Assis, Pesce, and Avanci (2006), the Wagnild and Young Resilience Scale is one of the few instruments used to assess levels of psychosocial adaptation in the face of significant life events, otherwise known as resilience. It considered 50 statements from a previous qualitative study with American woman identified as resilient considering (Wagnild & Young, 1990), from which the 25 items were written reflecting five dimensions (Wagnild, 2009)—self-reliance, meaningfulness, equanimity, perseverance, and existential aloneness—answered in a Likert-type scale. Results allow authors to find a bifactorial structure with good reliability and evidence of concurrent validity (Wagnild & Young, 1993) that has been studied in several countries and in cross-cultural studies (Hjemdal, Roazzi, Dias, & Friborg, 2015). In this sense, and recognizing the importance of resilience, this study serves to validate the psychometric resilience scale developed by Wagnild and Young for the Portuguese population based on a preliminary validation developed by Vara and Sani (2006) with some inconsistent data. Doubts about the internal structure of the measure in the Portuguese population must be overcome to use this measure in research and intervention studies.

METHODOLOGY

DISCUSSION

In the exploratory factor analysis of the Wagnild and Young Resilience Scale, the 10 most representative items loaded in a three-factorial structure: Factor 1—satisfaction with life—included four items (Young 16, Young 17, Young 21, and Young 6) with an eigenvalue of 2.195. Factor 2—planning-discipline—involved three items (Young 14, Young 15, and Young 1) with an eigenvalue of 1.991, and Factor 3—independence—is represented by Young’s Items 3, 5, and 2 with an eigenvalue of 1.748.

If we contrast the values obtained with the preliminary version of Vara and Sani (2006), we see that the factors obtained by us through the exploratory analysis are also focused on three factors, although their Factor 1 is made up of 14 items. We also see that both studies consider Items 16, 17, and 21, and that as far as Factor 2 is concerned, the preliminary version gives it 8 items, and Factor 3 is expressed in 4 items. So in this validation, none of the items obtained is concordant with those obtained by Vara and Sani. However, despite the divergence in the values presented from the mentioned validation scales, we infer that the Wagnild and Young validation scale, with the 10 variables assessed, was shown to be appropriate.

This procedure, with the elimination of some items to improve the validity of the measure has also been considered in other studies. Literature review allow us to find similar data in a paper from Damásio, Borsa, and da Silva (2011) in Brazil or Oliveira, Matos, Pinheiro, and Oliveira (2015) with a sample of Portuguese adolescents. Despite the existence of distinct definitions and measures, as we revised previously, these results might also highlight the need for further theoretical discussion and measure refinement. Some contributions to this debate can emerge from cross-sectional studies as the present one but also from cross-cultural studies where this same doubts emerge (Hjemdal et al., 2015). A tendency for the use of smaller measures, with items strongly directed to this construct with higher discriminant power should be considered (Aiena, Baczwaski, Schulenberg, & Buchanan, 2015; Losoi et al., 2013; Pritzker & Minter, 2014).

As far as the confirmatory factor analysis of the resilience scale was concerned, the main indicators of the model were analyzed (CFI, RMSEA, and the SRMR). We can infer that this is a noncentral chi-squared distribution whose complexity in our model is high (.988), that we have found an excellent fit index value for the model (.025), and that, as far as the SRMR index is concerned, our study attained a value of .035. In relation to the internal consistency of the three dimensions found, this was obtained through the Cronbach’s alpha coefficient in which we obtained a value of .82. Despite the elimination of the items, no factor maintains a Cronbach’s alpha lower than .7. Future studies should retest this solution and verify its impact of reliability of the measure. The development of new items considering this structure can be also considered.

As demonstrated and as result of the analyses carried out, we can conclude that the assessment of the Wagnild and Young Resilience Scale, through the 10 variables of the questionnaire, has shown itself to be appropriate. However, it is important to stress that, as Pesce et al. (2005) suggest, new studies will contribute to a better understanding and organization of the scale into factors, thus enabling further calibration and improvement of the instrument itself. It should be noted that, despite this instrument having presented suitable values for its validation, the main limitation of the study is considered to be the small size of the sample, which was made up mainly of a female population in a higher education institution. However, despite the constraints of the sample, this study is valued because we have obtained an adequate validation of the instrument, which can be translated into a better analysis and response on the interpretation of resilience for the Portuguese adult population.

In this sense, and considering the objectives of the study, it is considered that the Wagnild and Young Resilience Scale is valid for the Portuguese population. Future studies, with a broader and more representative sample, might cross-validate these results, allowing academics to know more about the resilience process in adults and its major predictors. Considering the stress and health challenges that nurses deal with, particularly pertinent research might be directed for a nursing context, exploring the role of inter- and intrapersonal characteristics predictors of resilience (Hart et al., 2014), but also organizational factors implied in this process (Cusack et al., 2016). Recent research also presented interesting results of intervention programs to promote resilience not only in new nurses (Chesak et al., 2015) but also in community settings (e.g., Dray et al., 2015; MacLeod, Musich, Hawkins, Alsgaard, & Wicker, 2016) that might be evaluated in our context with a measure with adequate properties. With sound measures, further interventions can be evaluated with deeper reliability.