Why externalize failure? Social goals, engagement and self-handicapping in the concern of eastern cultural community

Abstract

During this COVID-19 pandemic, the community, especially students, experienced anxiety due to the uncertainty that occurred. To cover it up, self-handicapping is the most common option. People in the eastern culture, such as Indonesia, prefer to achieve social achievement (SAch) goals than academic achievement (AAch) goals in motivation. This study aims to examine the model of the relationship between SAch goals with adaptive and maladaptive outcomes. This study used 904 respondents who filled out a questionnaire that had been tested for validity and reliability using confirmatory factor analysis and internal consistency respectively. The respondents of this research were young business practitioners who were studying in college. The results of the correlation test confirm the results of previous research that SAch goals consist of three interrelated but independent dimensions, so that students can pursue these three goals. Academic engagement also has two correlated dimensions, while self-handicapping has two uncorrelated and even contradictory dimensions. The results of testing the relationship model using a two-stage structural equation modeling approach further strengthen that behavioral self-handicapping was maladaptive. Social development goals (SDevGs) produced adaptive outcomes, while social demonstration goals (SDemGs) produced maladaptive outcomes. Eastern culture prefers did not make mistakes than showed social competence by appearing popular and having many friends. Need a curriculum that includes activities that encourage the achievement of social goals and feel an attachment to academic activities but minimizes behavior that externalizes failure.

Keywords:

• social achievement goals
• self-handicapping
• academic engagement

1. Introduction

In the context of academic learning, individuals often experience threats to their self-esteem. This threat is often posed by the fear of failure in a task or job that is considered important. According to self-worth theory, feelings of inadequacy can create shame (De Castella et al., 2013). These feelings lead to a desire to protect their self-esteem so that they trigger individuals to carry out self-handicapping strategies (Yildirim & Demir, 2020). Learning outcomes are influenced by various factors, such as goals, strategies, and motivation (Bembenutty, 2016). Among various self-protection strategies, self-handicapping stands out because it is widely known in academic settings (Ferradás et al., 2018). Self-handicapping is a preventive as well as defensive strategy against failure (Gadbois & Sturgeon, 2011). The strategy is carried out by individuals before achieving performance or carrying out certain activities (Gupta & Geetika, 2020).

Self-handicapping creates or claims obstacles to the achievement of success (Yildirim & Demir, 2020). There is agreement in the literature that self-handicapping is negatively related to processes and outcomes such as motivation and performance (Emadpoor et al., 2016; Phan & Ngu, 2014; Yu & McLellan, 2019). However, research results found that the relationship between self-handicapping and achievement varied, from insignificant, moderate, to strong negative (Schwinger & Steinsmeier-Pelster, 2011). Previous research tested self-handicapping as a defensive motivational strategy used by young people (Cocorada, 2011). Therefore, the strategy is negatively correlated with achievement (Javanmard et al., 2013; Rotairo et al., 2015; Torok et al., 2018; Zimmerman, 2013). Self-handicapping is based on attribution theory (Snyder et al., 2014), namely the reasons individuals behave. In the long term, self-handicapping affects the individual’s personality (Sahin & Coban, 2020).

Achievement goals have been proven to have a consistent effect on self-handicapping strategies (Urdan & Midgley, 2001; Yu & McLellan, 2019). These goals have indeed been widely used as a strong predictor of individual achievement (Shim et al., 2013). However, individuals also pursue other goals, such as social goals. This is because school is a complex environment that focuses not only on academic activities, but also on the development of interpersonal relationships (Goagoses et al., 2021; Goncalves et al., 2017). Some researchers have found that SAch goals are also associated with self-handicapping strategies (e.g., Cheng & Lam, 2013; Liem, 2016; S.S. Shim & Finch, 2014). SAch goals are related to various outcomes in the school environment, including academic achievement (Bardach et al., 2018). This is because the academic and social lives of students are closely related at school (Liem, 2016; S.S. Shim & Finch, 2014). Therefore, feelings related to others such as peers and the ability to adapt also affects adolescent behavior. According to Wormington and Linnenbrink-Garcia (2017), SAch goals are also related to outcomes such as engagement and well-being at school.

Besides being associated with academic success or failure, self-handicapping strategies are also often associated with academic engagement (Shih, 2013). This strategy is positively related to learning disengagement (Sawon et al., 2012). Studies on self-handicapping have been widely carried out in western societies with western cultures (Yavuzer, 2015). In the individualistic society, this strategy is used to protect private self-esteem, while in collectivist society, self-handicapping is used to protect public self-esteem (Pulford et al., 2005). Eastern societies with collectivistic cultures prioritize social goals over achievement goals (King et al., 2012). Community culture is an important factor because achievement can have different meanings in different cultures (Bahar et al., 2018).

The SAch goals pursued by students affect their behavioral and emotional engagement (Ryan & Shim, 2006). The dimensions of SAch goals have different characteristics. The SDevGs pursued by the students showed positive self-development. Most studies have shown that SDevGs predict behavioral and emotional engagement (Datu & Park, 2019). However, the pattern of the relationship is still diverse, there are those who report positive, negative, and some even report that they are not related to the two dimensions of engagement (Duchesne et al., 2019). Students who have positive social relationships will have engagement in quality learning and friendships (Bardach et al., 2018; Liem, 2016). Meanwhile, social demonstration avoidance goals (SDAvGs) have been consistently proven to be dysfunctional because they cause worry, anxiety, afraid of being evaluated and internalized behavior (Horst et al., 2007; S. S. Shim & Ryan, 2012). Various research results occur on the influence of social demonstration approach goals (SDApGs), namely on adaptive and maladaptive behavior (Bardach et al., 2018; Michou et al., 2016). Individuals who pursue this goal tend to rejoice in perceived popularity (Shim et al., 2013), but experience decreased personal growth and peer acceptance (Michou et al., 2016).

Research on the social aspects of academic motivation is still very rare but continues to grow until now (see, Berger & Archer, 2018; Giota & Bergh, 2020; Lee, 2018 for reviews). King et al. (2012) found that students’ SAch goals were related to behavioral and emotional engagement. The two dimensions of academic engagement can affect the self-handicapping strategy that is carried out. SAch goals encourage students to be engaged both behaviorally and emotionally. Self-handicapping will be chosen by students who have low academic engagement. This study aims to investigate the relationship between SAch goals, self-handicapping and academic engagement among students at several private universities. This study also examined the differences between males and females on the variables tested in this study.

2. Theory and hypothesis development

Self-handicapping is an important phenomenon that consists of several elements (Schwinger et al., 2014). Self-handicapping occurs when individuals feel uncertainty or doubt about their ability to excel (Snyder et al., 2014). This phenomenon occurs where individuals create barriers before evaluating their abilities. According to Clarke and MacCann (2016), these barriers can be positive (such as reducing anxiety, protecting self-esteem, increasing attribution and internal motivation) or negative (such as providing reasons or explanations for failure). Self-handicapping has been defined as the creation of barriers or weaknesses that interfere with optimal achievement in carrying out tasks (Zuckerman & Tsai, 2005). This self-handicapping strategy occurs unconsciously and begins when individuals doubt their abilities. The use of this strategy has negative psychological consequences in the long term, but increases self-esteem in the short term (Ommundsen, 2004).

Some of the behaviors used in self-handicapping strategies include lack of sleep, lack of preparation and effort, use of alcohol or drugs, and claiming to be sick (Petersen, 2014). This strategy is suitable for protecting self-esteem, developing excuses, externalizing failures, and reducing anxiety in the short term (Brown & Kimble, 2009; Petersen, 2014). However, previous research has found that self-handicapping results in long-term academic performance namely decreased academic achievement (e.g., Kwong & Davis, 2015; Schwinger et al., 2014) and individual personality (Üzbe & Bacanlı, 2015). However, the detrimental effect of self-handicapping on academic achievement was moderated significantly by the instrument used to assess this variable (Schwinger et al., 2014). Although it is still a debate, a review of self-handicapping also mentions that gender factors influence this phenomenon (e.g., Ferradás et al., 2018; Spinath et al., 2014). In addition, the age factor has also been reported to affect self-handicapping (Glazier et al., 2019).

According to social cognitive learning theories, maladaptive behavior such as self-handicapping mainly occurs in conditions of uncertainty of outcome and little hope for success (Torok et al., 2018). There are two forms of self-handicapping, namely self-reported handicapping or claimed self-handicapping and behavioral self-handicapping (Brown et al., 2012; Petersen, 2014; Warner & Moore, 2004). Self-reported or claimed self-handicapping are obstacles that are claimed to interfere with their performance or are claimed to have problems (Brown et al., 2012; Lee et al., 2021). Self-reported handicapping strategies include individual claims that they feel sick, anxious, in a bad mood, lack of sleep, or have experienced trauma in their lives (Yavuzer, 2015). Meanwhile, behavioral self-handicapping is when individuals actively or intentionally create obstacles to achieve success (Clarke & MacCann, 2016). The difference between the two is that behavioral self-handicapping actually occurs whereas self-reported handicapping is said or claimed to have occurred, although it does not necessarily occur (Clarke & MacCann, 2016; McCrea et al., 2008). Behavioral self-handicapping strategies include actions that directly impede performance, such as the creation of physical problems such as illness, injury, alcohol use, or unwillingness to study and being lazy to try (Lee et al., 2021).

Behavioral self-handicapping is more dangerous or maladaptive than self-reported handicapping because it provides fewer opportunities for success (Ferradás et al., 2018; Schwinger et al., 2014). Behavioral self-handicapping is more credible because it is more related to performance, namely lowering performance than self-reported handicapping (Schwinger et al., 2014). Students with high self-handicapping have high perceptions of test anxiety and emotional exhaustion, but low self-efficacy (Yildirim & Demir, 2020). Self-handicapping is related to learner characteristics such as motivation and attitudes towards learning activities (Thomas & Gadbois, 2007). Self-handicapping can be measured by means of observation or by self-report questionnaires (Clarke & MacCann, 2016).

Furthermore, self-handicapping strategies are much associated with achievement goals (Ferradás et al., 2018). Achievement goals are considered to play an important role in the selection of these strategies (Akin, 2014). Self-handicapping strategies are more positively related to performance goals than to mastery goals, while individuals who avoid appearing less competent are more likely to use these strategies to manipulate impressions of others (Thomas & Gadbois, 2007). The academic achievement goals model has been widely accepted (Korn & Elliot, 2016). There are not many studies that use SAch goals in academic settings, when in fact many students prioritize achieving social goals (Horst et al., 2007). College students spend a lot of time interacting with peers with different social aspects (Jones & Ford, 2014; Kiefer & Ryan, 2011). Although the number is increasing from time to time, there are still not many studies that use the influence of social goal motives on achievement (Ryan & Shin, 2011; Yu & McLellan, 2019).

Students’ academic and social life is strongly related to learning activities (Liem, 2016; S.S. Shim & Finch, 2014). Connecting with others, feeling belonging and accepted by peers, and fulfilling the desire to adapt can motivate students (Goagoses et al., 2021). SAch goals reflect the reasons why people engage in interpersonal behavior and achieve social competence (Ryan & Shim, 2008). SAch goals are the main reason for social involvement and how individuals are involved (Ryan & Shim, 2006). There are 3 SAch goals, namely SDevGs (development of peer relationships and improvement of social competence), SDApGs (showing social competence and getting favorable evaluations from others), and SDAvGs (hiding lack of social competence and avoiding negative judgments from others; Yu & McLellan, 2019). Previous studies have consistently found that SAch goals play a role in social adjustment and psychological well-being (e.g., Horst et al., 2007; Kuroda & Sakurai, 2011). Not many researchers have tested the effect of SAch goals on academic engagement (Goagoses et al., 2021; Shim et al., 2013).

Students with SDevGs are always associated with positive perceptions, no worries of being rejected by peers, focused on the learning process, adaptive, and high academic involvement (Liem, 2016). Furthermore, students with SDApGs are associated with high social competence but are worried about their behavior and social relationships (Ryan & Shim, 2006). These students who want to look cool and popular can underestimate their academic assignments, thereby lowering their academic engagement (Ryan & Shin, 2011). In other words, SDApGs are positively or negatively related to academic engagement. Meanwhile, SDAvGs are associated with negative perceptions of social relationships, worries, stress, and depression (Mouratidis & Sideridis, 2009). This can lead to low academic engagement of students (Shim et al., 2013). Individuals with SDApGs are more popular among students so that their self-confidence is higher than individuals with SDAvGs that have low self-confidence, worry, and are shy (Ryan & Shim, 2008). In general, SDevGs are more socially and psychologically adaptive than SDemGs.

One factor for motivation to influence academic performance is through academic engagement (Goagoses et al., 2021). Academic engagement is the involvement of students in schools in the form of participation, effort, interest, and enthusiasm (Skinner et al., 2009). There are two types of engagement, namely behavioral engagement (which is class participation by exerting energy, effort, and perseverance) and emotional engagement (which is an energetic emotional state such as enthusiasm, interest, and pleasure) that is felt in the classroom or in the learning process (Skinner et al., 2009). SAch goals can affect engagement in academic processes or activities adaptively (Goagoses et al., 2021). Achieving these social goals can increase behavioral and emotional engagement (Ramshe et al., 2019). Of the three SAch goals, only SDAvGs are associated with maladaptive consequences (Shim et al., 2013) and low engagement (Ben-Eliyahu et al., 2017). Meanwhile, SDevGs relate to prosocial behavior and satisfaction and well-being, while SDApGs relate to popularity and aggressive behavior (Shim et al., 2013). Students can pursue one or more SAch goals. Between the three SAch goals, the relationship is still very diverse, because few researchers have explored this relationship (Shin & Ryan, 2012).

The results of research by Ramshe et al. (2019) using university students as respondents found that SAch goals are a significant predictor of behavioral and emotional engagement. SAch goals relate to the reasons for involvement in an activity. King et al. (2012) found a relationship between SAch goals and the two dimensions of academic engagement. Peer support can strengthen emotional engagement (Fernandez-Lasarte et al., 2019), while SDAvGs have no effect on behavioral engagement (Hill & Wang, 2015). SDApGs are related to popularity, aggressive behavior and disengagement (Ben-Eliyahu et al., 2017). SDemGs are often associated with maladaptive outcomes such as self-handicapping (Shim et al., 2013). Implicit theory is a key factor that explains the relationship between academic self-handicapping and student disengagement (De Castella et al., 2013).

Cultural values can act as a deterrent to self-handicapping strategies. In Japan, for example, dependence on family, teachers and peers is high (Otsuka & Smith, 2005), so students do not use this strategy. This is because the negative consequences of self-handicapping outweigh the short-term benefits of externalizing failure. In the context of Asian society, adopting SDAvGs are less maladaptive (King, 2016; Liem, 2016). This study aims to examine the relationship between SAch goals, self-handicapping and academic engagement and examine the differences in gender in the three constructs. Relationship testing is done by testing the first-order construct, namely at the dimension level in each of these constructs.

3. Methods

3.1. Participants

This research was conducted at several universities in Yogyakarta, Indonesia. Yogyakarta is a student city in Indonesia where many students come to study in this city. The respondents of this research were business practitioners who studied at private universities in Yogyakarta. Henceforth, these business practitioners were called students. The students who were involved in filling out the questionnaire as respondents were those who are still actively learning, both offline and online. The students who filled out the questionnaire came from the economics and business faculties of several private universities in Yogyakarta. The data of this study were collected from questionnaires distributed to students by filling out google forms. Data collection was carried out for three months (August to October 2021). This data collection period was chosen after the second wave of COVID-19 began to subside in Indonesia. The link for filling out the questionnaire was distributed to 1500 students, but only 904 students filled out this research questionnaire (response rate 60.27%). They are between 21–24 years old, with 336 males and 568 females.

3.2. Measurements

This study used self-report questionnaires to assess the variables studied. This research was a research survey using a questionnaire adopted from several previous studies. Except for demographic variables, all question items used a-5 point Likert scale (1 = strongly disagree to 5 = strongly agree). The self-handicapping questionnaire was adopted from Clarke and MacCann (2016), for example, I will do a better job if I try harder (self-reported handicapping with a loading factor of 0.617 to 0.751 and α = 0.0.724), and I tend to delaying completing the task until the last minute or deadline (behavioral self-handicapping with a loading factor of 0.421 to 0.665 and α = 0.658). The SAch goals questionnaire was adopted from Bahar et al. (2018), for example, I feel successful when I learn something new about how to get along with other people (SDevGs with a loading factor of 0.573 to 0.806 and α = 0.905), it is important for me to appear to have many friends (SDApGs with a loading factor of 0.556 to 0.832 and α = 0.795), and it is important for me not to embarrass myself in front of my friends (SDAvGs with a loading factor of 0.503 to 0.760 and α = 0.857). Meanwhile, the emotional and behavioral engagement questionnaires were adopted from Van Ryzin et al. (2009), for example, college was fun (emotional engagement with loading factor 0.694 to 0.780 and α = 0.903) and I studied and worked hard on campus (behavioral engagement with loading factor 0.483 to 0.772 and α = 0.866).

3.3. Procedures

After the questionnaires were filled out by the students, validity testing with confirmatory factor analysis (CFA) and reliability with internal consistency was carried out to ensure a valid and reliable measuring instrument (Sekaran & Bougie, 2016). The validity test used factor analysis with a loading factor of more than 0.4 which means it was more important and internal consistency with a Cronbach alpha of at least 0.6 which means it was quite reliable (Hair et al., 2014). Before testing the relationship model using structural equation modeling (SEM), correlation testing was carried out to test the relationship between the variables used in this study (Hair et al., 2014). Furthermore, testing for gender differences in the variables studied used the independent sample t-test (Sekaran & Bougie, 2016). Testing the mediation of the relationship model was carried out using SEM with a two-step approach (Byrne, 2010).

4. Results

4.1. Descriptive statistics

Table 1 describes the mean, standard deviation, reliability with Cronbach’s alpha and the results of the correlation test between the first-order constructs used in this study.

Table 1. Descriptive statistics and relationships between research variables

	BSH	SRH	EEN	BEN	SDevGs	SDApGs	SDAvGs
Behavioral self-handicapping (BSH)	1.000
Self-reported handicapping (SRH)	0.005	1.000
Emotional Engagement (EEN)	−0.124**	0.541**	1.000
Bahavioral Engagement (BEN)	−0.462**	0.114**	0.264**	1.000
Social Development Goals (SDevGs)	0.027	0.661**	0.607**	0.105**	1.000
Social Demonstration-approach Goals (SDApGs)	0.203**	−0.044	0.024	−0.268**	0.004*	1.000
Social Demonstration-avoid Goals (SDAvGs)	0.110**	0.360**	0.365**	−0.030	0.413**	0.237**	1.000
Mean	2.867	4.184	4.038	3.423	4.154	2.698	3.727
Standard Deviation	0.618	0.644	0.673	0.714	0.669	0.734	0.785
Cronbachs’ Alpha (α)	0.724	0.658	0.903	0.866	0.905	0.795	0.857

**p < 0.01

*p < 0.05

Table 1 explains that the mean behavioral self-handicapping is lower than self-reported handicapping. This showed that students do not want to take actions that harm themselves and endanger their academic performance. They became self-handicappers because of the uncertainty and worry they feel. This uncertainty was related to the situational factors they face in the COVID-19 pandemic when filling out the questionnaire (Jia et al., 2020). Their enthusiasm and interest in learning materials was also higher than their actively involvement in the learning process. Of the three dimensions in the SAch goals, the average of SDevGs was the highest. This explains that students still want to improve their social competence. They also avoid bad judgment or mistakes from others, especially their peers. The mean of SDApGs was the lowest, which means the students do not need to be popular and do not want to be seen as having many friends.

The results of the correlation test showed that the three dimensions of SAch goals were significantly positively correlated (r = 0.04, p < 0.05; r = 0.413, p < 0.01; r = 0.237, p < 0.01). This showed that the construct was multidimensional which interacts even though the interaction was not very strong. In addition to the three dimensions of multidimensional SAch goals, the two dimensions of the academic engagement construct were also significantly positively correlated (r = 0.264, p < 0.01), which means that academic engagement is also multidimensional. Students’ interest and enthusiasm for learning were related to students’ active participation and involvement in learning, although the correlation was not strong. Meanwhile, behavioral self-handicapping and self-reported handicapping were not significantly correlated (r = 0.005, p > 0.05). The two dimensions of self-handicapping were independent and indeed different. This finding supports the results of previous research which states that behavioral self-handicapping is a behavior that has occurred, while self-reported handicapping is only claimed to have occurred (e.g., Clarke & MacCann, 2016; McCrea et al., 2008; Schwinger et al., 2014). Behavioral self-handicapping is also said to be more dangerous than self-reported handicapping (Brown & Kimble, 2009).

Table 1 also shows that behavioral self-handicapping was positively associated with SDApGs (r = 0.203, p < 0.01) and with SDAvG (r = 0.110, p < 0.01), but was not associated with SDevGs (r = 0.027, p > 0.05). This supports the results of previous studies that SDApGs and SDAvGs are associated with maladaptive consequences (Yu & McLellan, 2019). Furthermore, behavioral self-handicapping was significantly negatively related to emotional engagement (r = −0.124, p < 0.01) and with behavioral engagement (r = −0.462, p < 0.01). This finding supports the results of previous research which found that students who chose behavioral self-handicapping strategies were those who had low academic engagement or had disengagement (De Castella et al., 2013). Meanwhile, self-reported handicapping was significantly positively related to emotional engagement (r = 0.541, p < 0.01), behavioral engagement (r = 0.114, p < 0.01), SDevGs (r = 0.661, p < 0.01) and SDAvGs (r = 0.360, p < 0.01). Individuals with the goal of developing social competence and with the goal of avoiding making mistakes often experience a fear of failure. This is what causes them to claim that there are obstacles in their performance to protect their self-esteem (Petersen, 2014), even though the obstacles that are claimed are only concerns and do not necessarily occur. Table 1 also shows that self-reported handicapping does not correlate with SDApGs (r = −0.044, p > 0.05). This proves that the desire to look popular and have many peers is not related to worry or fear of failure.

Meanwhile, the correlation between emotional engagement and SDevGs as well as SDAvGs was significantly positive (r = 0.607, p < 0.01 and r = 0.365, p < 0.01), but not correlated with SDApGs (r = 0.024, p > 0.05). Students who want to develop their social competence had higher interest and enthusiasm. Those who avoid mistakes also had high enthusiasm and interest in academic activities, so their emotional engagement was also high. Furthermore, behavioral engagement was significantly positively correlated with SDevGs (r = 0.105, p < 0.01), significantly negatively correlated with SDApGs (r = −0.268, p < 0.01), but not associated with SDAvGs (r = −0.030, p > 0.05). SDevGs were consistently positively correlated with behavioral and emotional engagement, but social demonstration goals were not consistently correlated with these two dimensions of academic engagement. This finding supports the results of previous research, that SDevGs predict adaptive behavior, while SDemGs are not necessarily related to maladaptive behavior (Liem, 2016).

4.2. Gender difference test results

The results of previous studies found gender differences, especially in self-handicapping (Brown et al., 2012; Ferradás et al., 2018; Yavuzer, 2015). However, other researchers found no such difference (Leondari & Gonida, 2007). Table 2 presents the results of testing for gender differences in the variables currently being studied.

Table 2. Results of testing gender differences

	Gender	n	Mean	Standard Deviation	F	Sign.
Behavioral self-handicapping	Males	336	2.8168	0.6341	0.311	0.577
	Females	568	2.7920	0.6087
Self-reported handicapping	Males	336	4.2202	0.6276	0.819	0.366
	Females	568	4.1620	0.6534
Emotional Engagement	Males	336	4.0866	0.6744	0.425	0.515
	Females	568	4.0086	0.6716
Behavioral Engagement	Males	336	3.4071	0.7543	1.146	0.285
	Females	568	3.4317	0.6890
Social Development Goals	Males	336	4.2146	0.6630	0.199	0.656
	Females	568	4.1188	0.6706
Social Demonstration-approach Goals	Males	336	2.8447	0.8147	0.068	0.795
	Females	568	2.6112	0.7546
Social Demonstration-avoid Goals	Males	336	3.7656	0.7336	0.101	0.751
	Females	568	3.7038	0.7347

Table 2 is the result of testing gender differences in all the variables studied. The test results found that the average self-handicapping males were higher than females. Both males and females used self-reported handicapping more than behavioral self-handicapping. However, there was no gender difference in both self-reported handicapping and behavioral self-handicapping. In the academic engagement variable, both males and females have more interest and enthusiasm in learning or lectures instead of participating directly. This can be seen in the average behavioral engagement which was lower than emotional engagement, both males and females. The test results also showed that there was no gender difference in academic engagement. Furthermore, both males and females were more motivated for SDevGs than for SDemGs. The average of SDAvGs was higher than the SDApGs for both males and females. Similar with the other two variables, there was no gender difference in SAch goals.

4.3. Model mediation test results

Table 3 describes the results of testing the relationship model between SAch goals and self-handicapping mediated by academic engagement, while Table 4 describes the results of testing the relationship model between SAch goals and academic engagement mediated by self-handicapping.

Table 3. Test results of relationship model, academic engagement as mediator

	β	Critical Ratio
Social Development Goals → Emotional Engagement	0.605**	17.632
Social Development Goals → Behavioral Engagement	−0.049	−0.943
Social Development Goals → Self-reported Handicapping	0.654**	15.463
Social Demonstration-approach Goals → Emotional Engagement	−0.085**	−2.714
Social Demonstration-approach Goals → Behavioral Engagement	−0.315**	−8.280
Social Demonstration-avoidance Goals → Emotional Engagement	0.144**	4.010
Social Demonstration-avoidance Goals → Behavioral Engagement	−0.094**	−2.184
Emotional Engagement → Behavioral Self-handicapping	0.024	0.544
Emotional Engagement → Self-reported Handicapping	0.262**	5.812
Behavioral Engagement → Behavioral Self-handicapping	−0.601**	−13.297
Behavioral Engagement → Self-reported Handicapping	0.024	0.750
Emotional Engagement → Behavioral Engagement	0.417**	8.276
χ2 = 43.235 DF = 6 P = 0.000 GFI = 0.987 AGFI = 0.938 CFI = 0.978 NFI = 0.974 IFI = 0.978 TLI = 0.922

Table 4. Test results of relationship model, self-handicapping as mediator

	β	Critical Ratio
Social Development Goals → Self-reported Handicapping	0.811**	24.437
Social Development Goals → Behavioral Self-handicapping	−0.030	−0.587
Social Demonstration-avoid Goals → Self-reported Handicapping	0.127**	3.849
Social Demonstration-avoid Goals → Behavioral Self-handicapping	0.105**	2.001
Social Demonstration-approach Goals → Self-reported Handicapping	−0.116**	−4.024
Social Demonstration-approach Goals → Behavioral Self-handicapping	0.188**	3.933
Self-reported Handicapping → Emotional Engagement	0.759**	23.907
Self-reported Handicapping → Behavioral Engagement	0.244**	7.083
Behavioral Self-handicapping → Emotional Engagement	−0.227**	−6.982
Behavioral Self-handicapping → Behavioral Engagement	0.558**	−12.858
Social Demonstration-approach Goals → Behavioral Engagement	−0.215**	−5.886
χ2 = 38.082 DF = 7 P = 0.000 GFI = 0.987 AGFI = 0.950 CFI = 0.981 NFI = 0.977 IFI = 0.982 TLI = 0.944

Table 3 is the result of testing Model 1 where two dimensions of academic engagement mediated the relationship between SAch goals and self-handicapping. Model 1 fitted with the data because the values Goodness-of-Fit Index (GFI), Adjusted Goodness-of-Fit Index (AGFI), and Comparative Fit Index (CFI) are more than 0.9. Table 3 shows that SDevGs have a significant positive effect on emotional engagement, but not on behavioral engagement. SDApGs have a significant negative effect on emotional and behavioral engagement. Meanwhile, SDAvGs have a significant positive effect on emotional engagement but have a significant negative effect on behavioral engagement. The results of this study indicated that SApGs were predictors of academic engagement as in the results of previous studies (Goagoses et al., 2021; Liem, 2016; Ramshe et al., 2019).

Table 3 also shows that SDevGs and emotional engagement increase self-reported handicapping, while behavioral engagement can significantly reduce behavioral self-handicapping. The results of testing the mediation model using a two-stage approach to SEM showed that emotional engagement mediated the effect of SApGs on self-reported handicapping. In addition, behavioral engagement only mediated the effect of SDemGs on behavioral self-handicapping. This model was good enough because it has Normed-Fit index (NFI), Incremental Fit Index (IFI), and Tucker-Lewis Index (TLI) more than 0.90. Furthermore, Model 2 in which self-handicapping mediates the relationship between SAch goals and academic engagement is also tested in this study and is presented in Table 4.

Table 4 is the result of testing Model 2 where self-handicapping mediated the relationship between SAch goals and academic engagement. Model 2 fitted with the data because the values of GFI, AGFI, and CFI are more than 0.9. Table 4 shows that SDevGs have a positive effect on self-reported handicapping, but do not affect behavioral self-handicapping. SDAvGs can improve both dimensions of self-handicapping, while SDApGs can increase behavioral self-handicapping but reduce self-reported handicapping. SDApGs reduced enthusiasm and active participation of students in academic activities. Furthermore, self-reported handicapping increased both dimensions of academic engagement, while behavioral self-handicapping reduced both dimensions. The results of the mediation test of the model showed that SDevGs affected adaptive strategies, while social demonstrations affected maladaptive strategies. Furthermore, self-reported handicapping increased academic engagement, while behavioral self-handicapping caused disengagement. This model was also good enough because it has NFI, IFI, and TLI more than 0.90.

5. Discussion

This study examined the relationship between SAch goals, self-handicapping, and academic engagement among students at private universities in Indonesia, especially Yogyakarta. The results of this study indicated that behavioral self-handicapping and self-reported handicapping were not significantly correlated. This showed that the two dimensions were independent and reflective. The results of this test support the results of previous studies which found that the two dimensions of self-handicapping are indeed different (Brown & Kimble, 2009; Lee et al., 2021). Self-reported handicapping involved recognizing unfavorable conditions and giving external attributions to failure (Chen et al., 2009). Meanwhile, behavioral self-handicapping occurs when individuals create harm with their behavior which affects attribution and directly affects performance (Warner & Moore, 2004). In other words, behavioral self-handicapping creates self-weakness and attributes future failure, so it is more risky than self-reported handicapping which is only a verbalization of mistakes and worries (Ferradás et al., 2018).

Furthermore, the results of this study also found that the three dimensions of SAch goals were significantly positively correlated, but independent. This finding was important because students can pursue some of their SAch goals (S. S. Shim & Ryan, 2012). SDEvGs and SDApGs can be pursued together, so that the social competencies achieved are more positive than SDApGs that are pursued exclusively without any desire to develop social skills. Another interaction was that SDAvGs can be compromised as goals that do not necessarily lead to maladaptive outcomes when students were also pursuing the other two goals. The results of this study strengthen the findings of S. S. Shim and Ryan (2012) regarding the interaction of the three SAch goals.

In addition to the three dimensions of multidimensional SAch goals, the two dimensions of the academic engagement construct were also significantly positively correlated, which means that academic engagement was also multidimensional (Pilotti et al., 2017). The interest and enthusiasm of the students in learning were closely related to their perseverance and participation in learning. In the relationship model, interest, fun, and enthusiasm of students drove involvement, participation, efforts to avoid unethical behavior during learning, so that it prevented boredom and withdrawal from their school or campus.

The results of the gender difference test on the variables tested in this study found that there was no difference between males and females in self-handicapping, SAch goals, and academic engagement. This contradicts the results of previous studies which found gender differences in self-handicapping (e.g., Brown et al., 2012; Yavuzer, 2015) and SAch goals (Jones & Ford, 2014; S. S. Shim & Ryan, 2012). Both males and females prioritize self-reported handicapping and SDevGs. They also prioritized avoiding making mistakes when with their peers rather than showing off their popularity.

The results of the correlation test showed that the two dimensions of academic engagement had a significantly negative correlation with behavioral self-handicapping but significantly positively correlated with self-reported handicapping. Students who have enthusiasm, interest in learning materials, and enjoy learning activities will not do things that harm themselves such as not wanting to study or doing assignments. They worried that they will not be able to achieve the expected performance, so they tend to use a self-reported handicapping strategy. The results of this study supported the results of previous studies that behavioral self-handicapping was associated with disengagement (e.g., Clarke & MacCann, 2016; Goagoses et al., 2021; Petersen, 2014; Putwain, 2019; Stewart & De George-Walker, 2014; Yildirim & Demir, 2020). The results of this study support the results of previous studies which found that in an academic environment, behavioral self-handicapping was more emphasized than self-reported handicapping (Schwinger et al., 2014). Students’ self-reported handicapping can motivate them to try for achieving better academic performance, while behavioral self-handicapping can reduce their academic performance. The results of this study support previous findings, that individuals who perform behavioral self-handicapping are those who experience disengagement at school (Clarke & MacCann, 2016; Schwinger et al., 2014; Yu & McLellan, 2019).

Furthermore, SDevGs were significantly positively correlated with self-reported handicapping, but not with behavioral self-handicapping. Students who pursue the goal of increasing social competence will certainly not act that harms them. The results of this study support previous researchers who confirmed that SDevGs are associated with adaptive consequences such as prosocial behavior, sense of belonging, attachment, and satisfaction and social well-being (Shim et al., 2013; Yu & McLellan, 2019). Students choose the self-reported handicapping strategy because they want to protect their self-esteem. On the other hand, SDApGs were not correlated with self-reported handicapping but were significantly positively correlated with behavioral self-handicapping. Students who want to appear popular and appear to have many friends never worry about their achievements but are associated with aggressive behavior and disengagement. This encourages them not to be afraid to do things that harm them. This finding supports the results of previous studies, that SDApGs are associated with maladaptive behavior (e.g., Mouratidis & Sideridis, 2009; Shin & Ryan, 2012).

Meanwhile, SDAvGs were significantly positively correlated with both behavioral and self-reported handicapping. Students with SDAvGs have fears of making mistakes, but they also dare to do activities that make themselves not look wrong. This study confirms the results of previous studies, SDemGs are always associated with maladaptive consequences such as anxiety, avoidance, reduced self-efficacy and social well-being (Shim et al., 2013), low engagement (Yu & McLellan, 2019).

The results of testing the first relationship model showed that academic engagement mediated the effect of SDevGs and SDemGs on self-reported handicapping and behavioral self-handicapping. Meanwhile, the second relationship model showed that self-handicapping mediated the effect of SDevGs and SDemGs on self-reported handicapping and behavioral self-handicapping. SDevGs are proven to increase emotional engagement and self-reported handicapping, so they are proven to be adaptive. This confirms the results of previous studies (e.g., Datu & Park, 2019; Goagoses et al., 2021; Horst et al., 2007; Liem, 2016; Makara & Madjar, 2015; Michou et al., 2016).

SDAvGs were strong predictors of self-reported and behavioral self-handicapping, as well as reducing active involvement and participation in school (behavioral engagement). This is consistent with previous studies that found that SDAvGs were associated with maladaptive outcomes (e.g., Dawes & Xie, 2014; Michou et al., 2016; S. S. Shim & Ryan, 2012). However, the fear of making mistakes or appearing inadequate also motivated students to increase their interest and enthusiasm. This is consistent with the results of Liem’s (2016) study which found that avoidance goals such as SAch goals do not always weaken the motivation of people in collectivistic countries. The goal of avoiding negative social judgments encouraged individuals to adjust so that interpersonal relationships run smoothly and increase engagement.

Meanwhile, SDApGs reduced enthusiasm and interest, to active participation and involvement of students (emotional and behavioral engagement) and increased behavioral self-handicapping. This is consistent with the results of previous studies, that SDApGs trigger maladaptive behavior (e.g., Horst et al., 2007; Ryan & Shin, 2011; Shim et al., 2013). Surprisingly, SDApGs increased self-reported handicapping. This may be because the desire to be seen as popular or to appear cool or to gain the attention of others can trigger strategies for claiming inhibitions or wishing on hidden abilities.

6. Conclusion

Self-handicapping in students is not always caused by academic motivation. SAch goals have proven to be very useful in preventing self-handicapping behavior. Although not an academic motivation, SAch goals have been proven to increase academic engagement and reduce self-handicapping. SDevGs of students need to be improved by developing the quality of friendship and communication and interpersonal skills of students. This is in accordance with the conditions of society with a collectivistic culture. On the other hand, the desire to be popular or to be seen as having many friends is a maladaptive motivation in that culture. Meanwhile, the students’ worries and fears of making mistakes when together with their friends will limit their involvement in academic activities at school. This makes students behave adaptively or maladaptively.

SAch goals are more appropriate to apply to people with eastern culture. Therefore, teachers need to promote a learning environment that encourages students to believe in their abilities and focus on improving abilities, motivation, and academic engagement, as well as preventing self-handicapping. Because the three dimensions of SAch goals are related, it is hoped that SDevGs can reduce SDApGs that lead to aggressive behavior and SDAvG that reduce student academic engagement. In addition, curriculum development that includes social competence is needed for students to be able to increase academic engagement and reduce self-handicapping.

There are several weaknesses of this study that need to be known for future improvements. First, the data in this study were collected using a self-rated questionnaire, so there is a possibility of a common method variance, namely an increase in the value of the beta coefficient. Future research is expected to be able to collect data by different respondents separately between the dependent and independent variables. Second, the study uses cross-sectional data which causes the testing of the mediating variables to be less robust. Future research is expected to use longitudinal data collection techniques. Future research also needs to consider combining academic goals and social goals together.

Disclosure statement

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

Additional information

Funding

The author received no direct funding for this research.