The Brunner-Lei-Peters Resilience Scale

Resilience is an important attribute for mental health. Positive benefits felt by resilient individuals include fewer depressive episodes, and better coping strategies. Due to the positive outcomes experienced by resilient people, it may be clinically useful to examine this construct within a psychotherapeutic context. Accordingly, we created the Brunner-Lei-Peters Resilience Scale, which included a preliminary test of 70 items and 10 validity items; we tested the scale on a sample of 150 people. Participants were representative of the general population with about an equal representation of gender and a wide age range from 18 to 66 years of age. Although the scale initially included 70 items, through psychometric analysis, we reduced the scale to 15 items, while maintaining sound psychometric validity and reliability. Clinical implications and limitations of the scale are discussed. The Brunner-Lei-Peters Resilience Scale

10.29173/spectrum13  constructed our scale to be validated over a wide range of ages, be reasonably gender equivalent, and tap into more components of resilience.
As the current research only sampled adults (18+), we created resilience items geared towards assessment of an adult sample. Based on Taormina's (2015) insights, we included four subcomponents of resilience in adults: determination, endurance, adaptability and recuperability. Hence, many of our items assessed these four main dimensions (e.g., determination: "I am determined to succeed"; endurance: "I am empowered to succeed even in the presence of adverse circumstances"; adaptability: "I am adaptable"; and recuperability: "I recover from failures"). Although these four dimensions encompass a portion of resilience, we also hypothesized other components of resilience.
One element may be an ability to find humor in otherwise negative and serious situations. Kuiper (2012) uncovered that humor may help individuals as it leads to a high level of positive affect, which aids successful coping with trauma. In accordance with this research, we incorporated a few resilience items that asked participants about humor (e.g., "I can laugh at myself"). Another related component in resilience is the ability to utilize resources effectively to manage life's stressors. Prior research found that social and personal resources aided chronically-ill youth in coping with their adverse health; those without these resources were at risk for depression and a lower quality of life (Oleś, 2015). Thus, we also incorporated items that seemed to address participants' usage of available resources: "I am resourceful." Methods the survey. Ethnic breakdown is as follows: over half of the sample was Caucasian (66.7%); a substantial percentage of participants were Asian/Pacific Islander (31.3%); and the remainder of participants were Native American/Aboriginal (4.7%); Hispanic/Latino (2%); African American (0.7%) and Other (2%). Our percentages pooled to over 100% as we directed participants to choose all ethnic identities that applied to them. Ages ranged from 18 years to 66 years (M = 28.2, SD = 11.47). The mean age was right-skewed by a few older participants so we calculated the median age which was 23. This age better represents our data as the mode of participants (24.5%) were 21 years of age. Sixty percent of participants were assigned female sex at birth and 40% were male.
Gender identity for the sample was consistent, with 40% indicating their male gender identity and the rest citing female as their gender identity.

Materials
All three researchers independently formulated questions and then deleted duplicate items at a question-formation meeting. As some resilience scales already exist, we re-worded six items from two reputable resilience scales found online . In total, there were 80 items proposed, 10 of which were purposefully chosen as validity items. Although the 10 validity questions were not cross-validated, we used our psychometric judgment to compose high face-validity questions that directly assessed the resilience construct (e.g., "I am resilient).
We also reverse-scored 15 items as a means of deterring response biases (social desirability and acquiescence). Likewise, to minimize the effects of response bias, we informed students that their responses were anonymous, and made the title of our questionnaire purposefully ambiguous (Psychometric Questionnaire), rather than stating that the purpose was to study resilience. All 70 items plus the 10 validity items are included in

Procedure
Participants rated on a five-point scale how strongly they agreed/disagreed with each item from 1 (strongly disagree) to 5 (strongly agree).
Total scores were computed in Microsoft Excel for each participant. Validity score items were computed independently from the total scores of the scale. Each item's total score was separately correlated to the total validity score. By doing so, we eliminated items that did not correlate strongly to the total validity score measure (items with correlations less than 0.53). Our most strongly correlated items kept in the scale ranged from r = 0.53 "I have gotten stronger over time" to r = 0.64 "Even if I fail now, I can do better in the future." Before removing any items and after we reduced the scale to 15 items, we ran tests of reliability, validity, and conducted factor analysis using the Statistical Program for the Social Sciences (SPSS Version 23).
Scale was 75, and the lowest score possible was 15. The standard error of measurement for our scale was quite sensitive at 2.78.
As we collected demographic information from all participants, we also chose to examine whether different sexes, ethnicities, or age mattered in total resilience scores. A t-test revealed that there were no statistically significant differences between the resilience of males and females: t It is important to note that our median age of participants was 23, and age differences may be more likely if we had more participants of an older age. Finally, we also computed a One-Way ANOVA to test resilience between ethnic groups: F (4, 132) = 0.27, p > 0.89, which was inconclusive.

Reliability and Validity
Inter-item reliabilities of the 15 items utilizing Cronbach's alpha indicate strong internal consistency: α = 0.91. Our final analysis of items revealed no questions correlated to one another greater than 0.90, thus each question stood on its own and added separately to the variance.
Our items also have strong face validity as on the surface they appear to assess resilience.
Upon testing these items, construct validity was ascertained as r s = 0.86. We utilized Spearman's rank order correlation as our scale was rankordered and had equal intervals between rankings.

Cut Off Points
As seen through Figure 2, the line of best fit explains 79% of the total variance. We set our base rate score to 40/60 to diagnose only the top 1/3 participants as highly resilient. The lower-bound cutting line was placed at 30 as there appear to be three individuals who are not very resilient and are outliers compared to the rest of the group. The upper bound of the cutting score was placed at 60 out of a maximum of 75. Despite these outliers, most of the data is maximally clumped within a restricted range.
We tried to maximize the amount of variability accounted for by placing our cut-off points in optimal locations to minimize the amount of error. Research on resilience is important because the ability to bounce back after adverse situations contributes to positive psychological wellbeing (Mayordomo et al., 2016). For example, Tosun and Dilmac (2015), identified married individuals' levels of resilience were the second most important predictor in conflict resolution style, which is essential for marital success.
Research supports that clinical assessment should also focus on dynamic features clients can change rather than just static traits (Rutter, 2013).
Accordingly, one study uncovered that resilience may be adaptable (Wilson, 2016). Researchers taught students to practice gratitude as a means of bolstering resilience; those that practiced gratitude increased their ability to be resilient in a challenging educational setting. Although this research is in an educational domain, the study provides support that resilience is malleable and thus may be adaptable in therapeutic contexts too. Utilizing a scale akin to ours in a clinical setting may provide an initial baseline assessment of clients' self-perceived resilience. If patients score low on resilience, this provides informative cues about the extent to which they may succeed in psychotherapy.
The purpose of the study was to construct a psychometrically-sound scale of resilience. As resilience is indeed an abstract concept, we formulated items to best address the nomological concept of resilience. Since our questionnaire is short (15 questions), it is easily administered to get a quick view of how resilient clients views themselves. Clinicians may also wish to have others close to the patient fill out the questionnaire and see if there is a disjoint between the clients' perceptions and close others' views. The Brunner-Lei-Peters Resilience Scale is general enough to apply to all clients and doesn't focus on specific domains that one may be resilient within (e.g., health, academics, work). Accordingly, in therapy it may be beneficial to use the scale as an initial assessment tool, but also employ prompting questions to ascertain the unique resilient domains of each client. In sum, the Brunner-Lei-Peters Resilience Scale does not act as a substitute for therapy, but rather as a useful starting point.