1 An Exploratory Validation Study of a Measure of Other-Esteem with Adolescents

The purpose of this article is to provide further validation of initial exploratory results of a nascent self-report rating scale (citations blocked for review - authors 2017; author dissertation 2014) designed to measure the concept of other-esteem: The Other-Esteem Rating Scale (OthERS). The initial study provided preliminary norms, reliability, and validity for the OthERS with an undergraduate sample. To continue this line of inquiry, the OthERS was administered to adolescents ages 14–18. The article begins with a brief overview of the concept of other-esteem and the potential importance of researching the concept, followed by the methods used to further validate the OthERS, and concludes with suggestions for research and practice.

2 Other-Esteem

Other-esteem as conceptualized in this article was coined by Philip Hwang in Other Esteem: A Creative Response to a Society Obsessed with Promoting the Self (1995). Hwang stated that the impetus for the book was the belief that Western culture was over-focused on the self, independence, and self-sufficiency to the exclusion of consideration of other human beings. Hwang defined the concept: “Other-esteem is the respect, acceptance, caring, valuing, and promotion of all human beings, without reservation” (p. 14). Hwang’s (1995) theoretical conception comprises 10 aspects of other-esteem: Non-offensiveness, friendliness, courteousness, kindness, respectfulness, acceptance, valuing, praising, promotion of others, and ability to forgive.

· Non-offensiveness refers to not engaging in physical or verbal abuse of others.

· Friendliness refers to being amicable and social.

· Kindness is helping others in need.

· Respectfulness is showing regard for someone else.

· Acceptance is the ability to acknowledge and welcome others’ individual, social and cultural uniqueness.

· Valuing is treasuring a relationship one has with another person.

· Praising is cheering others on.

· The promotion of others is when a person helps another to achieve new goals and accomplishment without jealousy or competition.

· Forgiveness is the ability to forgive a person who has hurt you.

Other-esteem was searched in Academic OneFile, EBSCOhost, Proquest, LexisNexis, and ERIC. Hwang was the only other result found regarding other-esteem as conceptualized in this article. The article now turns to examine selected hypothetical correlates of other-esteem that impact people’s lives, namely bullying and aggression.

3 Bullying

The U.S. Department of Health and Human Services defines bullying as unwanted, aggressive behavior that involves a power imbalance, whether real or perceived. Bullying has been described as having three key elements: An intention to harm, repetitive in nature, and a power imbalance between the perpetrator and victim (Felix et al., 2011; Olweus 1993; Rodkin et al., 2015; Werth et al., 2015). Bullying can occur in many different settings including in the school, community, work, home and online.

Of particular concern to those studying bullying is the myriad possible effects it can have on the victim and perpetrator. Some research has indicated both victims and perpetrators are at increased risk of suicidal ideation and suicide attempts (Holt et al., 2015; Klomek et al., 2009). Other adverse effects can include poor mental health, depression, and social maladjustment (Hawker & Boulton, 2000; Moore et al., 2017; Ttofi et al., 2011), poor academic achievement (Nakamoto & Schwartz, 2010), and illicit drug use and substance abuse for victims (Doumas et al., 2017; Moore et al., 2017).

The National Center for Educational Statistics (NCES, 2017) reports one out of every five students (20.8%) report being bullied. Of those who reported being bullied, 33% were bullied at least once or twice a month during the school year; 64% were made fun of, called names, or insulted; 59% were the subject of rumors; 24% were pushed, shoved, tripped, or sat on; and 24% were excluded from activities on purpose. Modecki and colleagues (2014) found that rates of self-report bullying varied between 9% and 98%.

Due to the possible high prevalence of bullying and the effects bullying can have on victims it is important to examine possible correlations between personality measures and bullying behavior. This exploration could lead to better ways of identifying individuals at risk for bullying behavior, as well as create interventions to bolster positive traits associated with lower risk for bullying behavior. Multiple personality traits have been theorized to be associated with bullying behavior including low self-esteem and empathy.

4 Self-Esteem, Empathy, and Bullying

Branden (1969) theorized that self-esteem affects every moment of an individual’s existence. Branden stated that an individual’s self-evaluation is the context in which individuals sets goals and meet the challenges of life, and further, similar to both Maslow and Erikson, that when self-esteem is stunted, psychological development becomes stunted. According to Branden individuals with low self-esteem inevitably allow the negatives in their life to have more power than the positives, thus creating an inability to find joy in their accomplishments, and that more serious psychological conditions follow if a person has severely low self-esteem.

Researchers have examined potential relationships among self-esteem, aggression, and bullying. The correlation between bullying behavior and self-esteem is a controversial topic.

O’Moore and Hillary (1997) found that children who engage in bullying behavior had lower self-worth, and Byrne (1994) and O’Moore (1997) both reported lower self-esteem in children who engage in bullying behavior. Bushman & Baumeister (1998) conducted laboratory experiments on. aggression and self-esteem. High or low self-esteem did not predict aggression in. this study. Jang & Thornberry (1998) followed nearly 1000 students for 4 ½ years to. examine the relationship between delinquent behavior and self-esteem. The authors. found a small to negative correlation between high or low self-esteem and delinquent. behavior (r = − .22). Wang et al., (2013) examined self-esteem and aggression in 8085 adolescents in Taiwan. The researchers found no significant difference in self-esteem among perpetrators of bullying and neutrals. Thomaes et al., (2009) examined self-esteem buttressing. The researchers stated, “Interventions aimed at buttressing self-esteem lessen the psychological impact of ego threat by focusing the individuals on the core traits that define them as a person. Such interventions do not artificially raise or deflate self-esteem” (p. 1536). The results of these studies are mixed. The relationship between bullying behavior and self-esteem is not clear.

Similarly, research on the relationship between empathy and bullying behavior is mixed. Some researchers (Joliffe & Farrington, 2011; Munoz et al., 2011) found a positive correlation between low empathy and bullying behavior, whereas other researchers (Stanbury et al., 2009; Vachon et al., 2013) found no correlation between low empathy and bullying. Due to the mixed results, researchers have begun to study empathy as a predictor of willingness to intervene on behalf of a bullied peer (Walters & Espelage, 2019) - it is unclear if a relationship between high empathy and the willingness to intervene on behalf of a bullied peer exists. Schultz et al., (2014) stated that this outcome may be because the definition of empathy is too broadly defined. Similar to empathy, emotional intelligence has failed to correlate significantly to bullying or aggressive behavior (Castillo et al., 2013; Kokkinos & Kipritsi, 2012; Mavroveli & Sanchez-Ruiz, 2011).

The research on self-esteem, empathy,and bullying indicates a gap in the literature. Although they might theoretically be expected to be linked, none of these constructs have been found to be strongly related to aggression or bullying at this time.

5 Potential Importance of Other-Esteem

During the initial validation study it was found that the OthERS was moderately negatively correlated to aggression. This correlation was higher than any relationship between bullying or aggression with high self-esteem, low empathy or low emotional intelligence reported in the literature (Busse & Flowers, 2018). Other-esteem may be a concept that helps to screen for individuals at risk for bullying behaviors and aggression, and to provide interventions and support. Further, perpetrators of bullying behavior also may have negative outcomes (Holt et al., 2015; Klomek et al., 2009). Identifying and supporting individuals at risk of bullying behavior may be an effective way to address some of the negative outcomes associated with bullying and other aggression.

6 Current Study

The purpose of the current study was to explore the potential validity of the OthERS with individuals aged 14–18, and to evaluate the relationships among self-esteenm, aggression, and bullying with other-esteem. The overall goal of this line of inquiry is to develop a concept which might help to predict future bullying or aggressive behavior more reliably than the constructs of self-esteem orempathy. Reliability and factor analyses were utilized to examine the structure of the OthERS with an adolescent sample.

7 Method

7.1 The OthERS

Using Hwang’s concepts, items were created and placed on a 4-point response scale (never, sometimes, often, always). The initial version of the OthERS consisted of 27 items after content validation by five psychology professors with expertise in test construction and social-emotional assessment. The OthERS was initially validated with undergraduate students but was created as a child/adolescent measure. The OthERs was created to address the gaps in the literature in regards to finding a.

Further content validation was accomplished through item discrimination analyses. Item discriminations (or biserial correlations) describe how well an item unambiguously classifies an item as “pass” or “fail” (DeVellis, 2017). Researchers examine how an item performs against the total scores of the subjects. When an item discriminates well it has a narrow range of ambiguity in regard to what is being measured. If an item has weak discrimination power, then it is not adding to the total understanding of what is being measured and should be eliminated, Items with discriminations under 0.3 were eliminated from the final measure (DeVellis, 2017).

The final scale of the OthERS was normed with 227 undergraduate students (citation blocked for review, 2017). The total scale alpha was 0.89, indicating strong internal consistency, but temporal stability was weak (r = .53). The test-retest data were gathered from a small, convenience sample during university final evaluation time, which may have influenced the data. Conversely, it may be that the construct is not stable across time.

The OthERS evidenced a moderate negative correlation with aggression (r = − .54), and a moderate positive correlation with self-esteem (r = .35). An oblique exploratory factor analysis supported five factors that accounted for 60% of the variance. These five factors were dubbed: Praise and Kindness, Respect for Others, Forgiveness, Social Etiquette, and Social Responsibility. The OthERS was administered online, along with the Buss-Perry Aggression Questionnaire-Short Form (AQ) (Buss & Perry, 1992), the Forms of Bullying Scale-Perpetration (FBS-P) (Shae et al., 2013), and the Rosenberg Self-Esteem Scale (RSES) (Rosenberg, 1965). This administration was conducted through the online survey program Qualtrics and responses gathered through Pollfish, an online tool for soliciting participation in research studies. Pollfish typically compensates individuals between $1.00 and $3.00 to complete a survey. Reliability and validity were examined via Jamovi, an online analysis tool (The Jamovi Project, 2019). It was predicted that the OthERS would evidence small to moderate inverse correlations with bullying and aggression, and a small, positive correlation with self-esteem, as found in the previous study with undergraduate college students.

7.2 Buss-Perry Aggression Questionnaire-Short Form (AQ)

The AQ (Buss & Perry, 1992) is a 29-item five-point self-report measure designed to assess four areas: physical aggression, verbal aggression, anger, and hostility. Bryant & Smith (2001) created the shortened version of the AQ through a factor analysis study with three multinational sample populations from previous research which yielded a 12-item measure with internal consistency estimates ranging from 0.70 (hostility) to 0.83 (verbal aggression). The Cronbach’s alpha with the current sample was 0.90 (omega = 0.90).

7.3 Forms of Bullying Scale-Perpetration (FBS-P)

The FBS-P (Shaw et al., 2013) is a 10-item bullying scale that asks behavioral questions about bullying behaviors on a five-point scale: This did not happen to me/I did not do this; Once or twice; Every few weeks; About once a week; and Several times a week or more. The FBS-P was normed on 3,496 8th grade students, and 783 8th -10th grade students.

Shaw et al., (2013) reported two separate studies validating the FBS-P. The FBS-P was normed on 3,496 8th grade students in Australia. The FBS-P displayed strong internal consistency (Cronbach’s alpha: 0.87) and a one-factor structure. The FBS-P positively correlated with a measure of conduct disorder and negatively correlated with a measure of pro-social behaviors. In the second study, the FBS-P was administered to 783 8th -10th grade students in Australia. Factor analysis supported the single factor structure. The FBS-P evidenced strong reliability (Cronbach’s alpha: 0.91) and was found to correlate with measures of conduct disorder and pro-social behaviors in a manner similar to the first study. Cronbach’s alpha with the current sample was 0.93 (omega = 0.93).

Though no validation studies were found on the FBS with individuals aged 14–18 the FBS was chosen for the following reasons: it has a definition of bullying with a time element (i.e., happened this year, happened last week), includes both physical and verbal bullying and was developed by utilizing the definition of bullying from Ortega and colleagues (2001) which includes intention, repetition, and power imbalance.

7.4 Rosenberg Self-Esteem Scale (RSES)

The RSES (Rosenberg, 1965) is a 10-item self-report rating scale that is widely used in research (Wongpakaran & Wongpakaran, 2012). The scale was developed with 5,024 high school students in New York State, USA. The internal consistency of the RSES was reported as 0.77 (Dobson et al., 1979), and 0.88 (Fleming & Courtney, 1984; Silber & Tippett, 1965) reported a test-retest correlation of 0.85 for 28 participants after a two-week interval and Fleming & Courtney (1984) reported a correlation of 0.81 for 259 participants with a one-week interval. Cronbach’s alpha with the current sample was 0.77 (omega = 0.77).

Savin-Williams & Jaquish (1981) found the RSES correlated 0.72 with the Lerner Self-Esteem Scale, which demonstrated adequate convergent validity. Supple et al., (2012) examined the factor structure of the RSES with 1,248 European American, Latino, Armenian and Iranian adolescents. Internal consistencies ranged between 0.79 and 0.89 across the different cultural groups. The researchers found that the RSES two factor structure (positive self-esteem and negative self-esteem) was consistent across the cultural groups.

8 Ethics

A waiver of parental consent was requested for this study. Using a survey program such as Pollfish (2019) allows for research to quickly survey minor individuals aged 14–17, but does not provide a method for gathering parental consent. Requiring parental consent would limit a researcher’s ability to gather anonymous data by asking the participants for a parent email or contact information. Secondly, the FSB-P asks sensitive questions about bullying behavior such as: “I deliberately physically hurt or ganged up on someone,” and “I deliberately damaged, destroyed and/or stole someone’s things.” Answering these questions could place participants at risk for discipline or criminal consequences if parents were notified. Further, if identifying information were required, participants may not answer truthfully or may opt out of the study to avoid perceived consequences. Current federal regulations allow for the waiver of parental permission (Protection of Human Subjects, 2019, pt. 408.c). Based on these regulations and recommendations, the Institutional Review Board waived parental consent for the current study.

9 Participants

A convenience sample of adolescents aged 14–18, was recruited in order to evaluate the structure of the measure and correlates with adolescents at the ages the measure is most likely to be used in practice. Adolescents younger than 14 were not able to be sampled, as the sampling tool used, Pollfish, only provides respondents aged 14 and up. The participants were 486 individuals who completed a short demographics questionnaire, the OthERS, AQ, FBS-P, and RSES. Of the total sample, 166 were female (34.4%) and 306 (63.4%) were male with a mean age of 16.36 (range 14–18); 307 (63.7%) self-identified as White, 67 (13.9%) as Black, 21 (4.3%) as Asian, 46 (9.5%) as LatinX, and 42 (8.7%) self-identified as Other; 166 (34.4%) identified as living in the northeast, 115 (23.8%) in the Midwest, 139 (28.8%) in the South, and 63 (13.0%) identified as living in the West. See  1 for additional demographic information.

Table 1 Participant demographics
Full size table

10 Final Scale

One item was eliminated due to confusing wording which resulted in a 21-item scale, with a potential raw score range of 21–84. Based on the revised scale, the participants’ mean score was 60 with a standard deviation of 8.81, mode of 60, and a median of 60. A score of 55 on the OthERS placed an individual in the 25th percentile, 60 was at the 50th percentile, and a score of 66 was at the 75th percentile. Therefore, respondents in this sample who scored below 55 may be considered as having lower other-esteem and those who scored 66 or above may be considered as having higher other-esteem.

11 Validity

Construct and criterion validity were examined. Exploratory factor analysis (EFA) was used to assess construct validity, and correlations between the OthERS and RSES were used to assess criterion validity. Additional correlations were examined between the OthERS and the BP-AQ and the FBS to examine the potential treatment utility of the OthERS (i.e., if it could be used as a screening measure to identify those at-risk of bullying).

11.1 Construct Validity

To determine if the data were adequate for factor analysis, two standard techniques were used. The Kaiser-Meyer-Olkin (KMO) test should be at least 0.70 and Bartlett’s test of sphericity should be significant at the p ≤ .05 level for a data set to be considered appropriate for factor analysis. The KMO result was adequate (0.925) and Bartlett’s test of sphericity was significant, χ2 (210) = 4956, p < .001), indicating the sample and data were adequate for factor analysis.

A previous EFA was conducted with college-age students, and results indicated the scale fit best within a five-factor model (citation redacted for review, 2018). There is a tendency for EFA to over-fit a sample, thereby making it difficult to replicate a structure in an alternative sample (Osborne, 2014). Replication of an EFA structure is particularly unlikely when the alternative sample differs from the original in some potentially meaningful way. The current sample differed from the original sample in age, gender distribution, sample size, and other characteristics. Given the differences between the sample in the current study and the original validation study (e.g., age and gender distribution), as well as limited research establishing the structure of the OthERS in alternative samples, it was determined a replication EFA was more appropriate than a confirmatory factor analysis (see DeVellis 2017; Osborne, 2014).

Total scores and all individual items on the measure had non-normal distributions (see Table 2 for the distribution of frequency counts across response options for each item), indicating a maximum likelihood extraction method was inappropriate (Fabrigar et al., 1999). Therefore, a principal factors extraction method with promax rotation was utilized, considering the potential factors were assumed to be related. Parallel analyses were used to determine how many factors to include in the final model (Fabrigar et al., 1999). Four factors emerged that accounted for 54.1% of the variance (see Table 3 for additive and cumulative variance). Although lower than 60%, this is a minimally acceptable structure to explain the variance of the OthERS with this sample of adolescents. Fit indices indicate a good fit of the factor structure to the data (RMSEA = 0.050 [90% CI: 0.042 − 0.058]; X2(132) = 290, p < .001). The four factors evidenced small to moderate inter-scale correlations (see Fig. 1).

Table 2 Frequency Counts of Item Responses on the OthERS
Full size table
Table 3 SS Loadings and Percent of Variance
Full size table

Fig. 1: EFA four-factor model. Factor correlations are represented by curved lines, with r reported for each. Straight lines represent factor loadings, with the factor loading coefficient reported beneath the line for each item

Factor one is comprised of eight items. This factor was named Praise. Factor two is comprised of five items. This factor was named Acceptance. The third factor is comprised of four items and was named Respect for Others. The fourth factor is comprised of four items and was named Forgiveness. This model is different from the original Others EFA structure in number of factors (4 vs. 5) and percentage of variance accounted for (54.1% vs. 60%). Factor item loadings can be found in Fig. 1.

Fig. 1
figure a

EFA four-factor model. Factor correlations are represented by curved lines, with r reported for each. Straight lines represent factor loadings, with the factor loading coefficient reported beneath the line for each item

Full size image

11.2 Concurrent Validity

The total OthERS score was compared to a measure of self-esteem (RSES). There was no correlation between the OthERS and the measure of self-esteem (r = -.102).

12 Reliability

Reliability of the final scale was evaluated based on estimates of internal consistency. Cronbach’s alpha was used due to the multiple response format. The internal consistency of the total 21-item scale was examined. The resulting Cronbach’s alpha was 0.85 (omega = 0.88), which indicates the total scale possesses good internal consistency. Additionally, the four factors identified through the EFA evidenced adequate internal consistency: Factor 1 (Respect for Others) alpha = 0.90 (omega = 0.90), Factor 2 (Acceptance) alpha = 0.79 (omega = 0.83), Factor 3 (Praise) alpha = 0.82 (omega = 0.82), and Factor 4 (Forgiveness) alpha = 0.77 (omega = 0.77).

13 Aggression and Bullying Comparisons

The OthERS was compared to the BP-AQ and the FBS. Residual plots were examined for linearity, and the relationships appeared to be mostly linear, although the relationships between the OthERS and the FBS-P had a slight curve. The OthERS evidenced a small inverse correlation (r = -.29, p < .001, r2 = 0.09) with the FBS-P, indicating that low other-esteem may be a predictor of bullying behavior. Similarly, the OthERS showed a small inverse correlation (r = -.19, p < .001, r2 = 0.04) with aggression.

Additionally, given our findings indicated a relationship between self-esteem and bullying and aggression, we compared the correlations between the OthERS and the FBS-P to the RSES and FBS-P, as well as the correlations between the OthERS and the BP-AQ to the RSES and the BP-AQ by converting each r-value to a z-score and calculating a z-test statistic for each comparison. On the FBS-P, the OthERS correlation was significantly greater than the RSES (zobserved = 2.06, p < .05). There was no statistical difference between the OthERS and RSES on the comparison to the BP-AQ (zobserved = 1.14). (See Table 4 for correlations between each measure.)

Table 4 Correlations among the Validation Measures
Full size table

14 Age and Gender Comparisons

Participants had three gender options: male, female, and other. Comparisons indicate no differences on total OthERS scores between any gender selection. Participants ranged in age from 14 to 18. Differences, corrected for multiple comparisons, between each age group were analyzed. Means and standard deviations for each age group were visually similar, except for those who were 18: 14 (M = 60.7, SD = 9.31), 15 (M = 61.4, SD = 10.9), 16 (M = 60.7, SD = 8.95), 17 (M = 60.5, SD = 8.93), and 18 (M = 53.7, SD = 11.7). Significant differences were found between participants who were 15 and 18, t(478) = 3.74, ptukey = 0.002, d = 0.68), those who were 16 and 18, t(478) = 4.17, ptukey < 0.001 d = 0.67, and those who were 17 and 18, t(478) = 3.91, ptukey < 0.001, d = 0.65. No additional differences were found between age on total scores.

15 Discussion

This study aimed to validate the OthERS with an adolescent sample with ages ranging from 14 to 18. The OthERS evidenced good internal consistency. The anticipated relationship between other-esteem and self-esteem was not observed in this study. Inverse correlations were observed between both bullying and aggression and the OthERS. The relationships found between the OthERS and aggression and bullying in the current study are stronger than both the small relationship found between self-esteem and bullying and aggression in the previous study, as well as the reported correlations in the literature between self-esteem, empathy or emotional intelligence and bullying behavior. Concerns about the nonlinearity of these relationships have been discussed elsewhere (Perez et al., 2005; Webster, 2007).

An EFA was conducted to examine the factor structure with the sample of younger adolescents in the current study. The factor structure of the measure changed from the initial measure validation study with a college-age population. Six items that were included in the initial study were excluded in the present study due to low factor loadings. With a sample of college students, the OthERS evidenced a five-factor model, but with adolescents ages 14–18 a four-factor model emerged. This result may be due to developmental differences between the two populations of middle vs. late adolescence (note: there is no consensus on what constitutes the age ranges of early, middle or late adolescence), or to the tendency of EFA to over-fit a sample (Osborne, 2014).

16 Limitations

A major limitation of this study is that participants were recruited through an online third-party survey program. Wright (2005) discussed the major limitations of this method. The first is sampling issues, given very little is known about the characteristics of the individuals completing online surveys. Another issue with online survey sampling is self-selection bias. Also, the participants were compensated for completing surveys, so the participants may be more likely to be motivated by monetary reward. Further, survey participants need to have access to an electronic device, as well as the internet.

In addition, the sample was not stratified so as to be representative of gender, age, and race/ethnicity. For example, 63% of the respondents identified as male, and ages 14, 15, and 18 were underrepresented. Differences on the OthERS between gender and age selections were examined. No differences in responding between genders were found. Means and standard deviations across ages were similar, aside from 18-year-olds, whose mean scores were significantly different than 15-, 16-, and 17-year-old respondents. On average, 18-year-olds scored lower than other ages on the OthERS, indicating lower other-esteem. There are many possibilities for this finding, but some may be that as we age, we have less of an inclination to engage in response bias, may have improved self-awareness of actual other-esteem, or there may be a possible developmental decline in other-esteem as we age.

Another limitation relates to the use of self-report measures, particularly regarding bullying and aggression. Social desirability bias (Althubaiti, 2016; Randall, 1991) and recall bias (Althubaiti, 2016; Bell, 2019) is a common problem in self-report survey research. Participants may report themselves as overly positive and may not recall information accurately. Whereas these limitations should not be discounted, the results reported here are similar to previous findings, have adequate-to-strong internal consistency, and were found across measures. Further, few differences were found across age and gender, and no significant threats to validity were present.

17 Implications for Practice and Future Research

The current study has shown that the OthERS possesses moderate to strong psychometric properties and provides additional evidence for empirical study of the concept of other-esteem. Further study is warranted with a representative sample of mid- to late-adolescents that is stratified according to the current census to replicate and further explore factor structure.

One limitation of the current study was the lack of verification of results, as all measures were collected only via self-report. To eliminate this problem in the future, a parent and teacher report version of the OthERS may be developed, so that researchers and practitioners can assess other-esteem through multiple sources.

A related area for research is to extend validation studies of the OthERS to clinical populations hypothesized to have diminished other-esteem, such as those with conduct disorders and antisocial personality disorder. Other-esteem could also be examined in relation to life satisfaction. In the initial validation study with college-age students, life satisfaction was moderately related to other-esteem. Finally, other-esteem could be examined across the lifespan and across cultures to observe differences over time and across populations.

Practitioners might use the concept of other-esteem in a strength-based approach. The OthERS could be used to help to identify areas of social strength for a student and this could be used to create opportunities for success in the school setting. For example, when a student is exhibiting low other-esteem a practitioner could use the OthERS to identify specific areas to strengthen and to create an intervention plan to bolster the student’s other-esteem.

Additionally, other-esteem fits well within the realm of social-emotional learning (SEL), which has the primary goals of developing self-awareness, self-management, social awareness, relationship skills, and responsible decision-making (CASEL, 2020). These goals have significant overlap with other-esteem, particularly in the areas of social awareness and relationship skills. Given this overlap, it is possible the OthERS could be utilized as a screening or progress monitoring tool within SEL curricula. In this way, results on the OthERS could identify students with low other-esteem who may be in need of additional supports within tiered intervention. Depending on individual item responses, or scale scores, results could inform educators on potential SEL target skills and tiered interventions could be developed to support students’ SEL in a MTSS.

18 Conclusion

Other-esteem may fill a gap left by self-esteem to address socially valid components of esteem, with some predictive relationships for difficult-to-predict behaviors, such as bullying and aggression. The results of this exploratory study with adolescents ages 14–18 have provided further evidence for the potential benefit of measuring other-esteem. The OthERS evidenced adequate internal reliability and good initial construct validity. As hypothesized, the OthERS was inversely related to measures of bullying and aggression, indicating it may have some level of treatment utility in bullying prevention, although further research is needed in this area to validate this finding. A measure of other-esteem may provide direction for theory, research and practice. At present, the OthERS is a nascent tool in need of further validation. We invite debate and exploration of this construct.