Analyzing the Commitment of College Students Using a Brief,Contextualized Measure of Need Satisfaction From the Perspective of Self-Determination Theory

Abstract

This study empirically confirmed the relationships between the degree to which students satisfied three basic needs (competence, relatedness, and autonomy) and the strength of their commitments to the university they attended and to obtaining a baccalaureate degree. A questionnaire was administered online to 1257 students at two 4-year universities. Regression analysis yielded statistically significant associations between the three needs and Institutional Commitment and Degree Commitment, explaining more than 20% of the variance in the latter two variables.

Keywords

Degree Commitment Institutional Commitment self-determination theory basic needs

Introduction

The college years are often a time when great strides are made in students’ personal, social, and intellectual growth (Pascarella & Terenzini, 2005). And certainly institutions of higher education are structured and organized to foster the students’ growth and development (for review, see Tinto, 2012). However, numerous indicators confirm that many students flounder or fail to fulfill their potentialities. The correlations between measures of academic abilities (such as aptitude tests) and college grades are weak to moderate (<less than .37; Robbins, Lauver, Le, Davis, & Langley, 2004), and only about 60% of students who seek a bachelor’s degree or its equivalent attain it within six years (National Center for Education Statistics, 2016; Shapiro et al., 2015).

Cognizant of these discouraging statistics, higher education schools have devoted substantial attention and resources into initiatives aimed at improving the social and academic climate. Ones with high retention rates often serve as models of best practices, and their techniques are widely circulated for others to emulate (American College Testing [ACT], 2010). The national picture, however, remains grim in spite of the salutary efforts of many, if not most, postsecondary schools to incorporate the latest ideas. In the words of one highly influential expert:

However one analyzes the data, one fact remains clear. Institutional rates of four-year and two-year degree completion have not increased substantially since the 1990s. … Despite years of effort, institutions have yet to develop a coherent framework to guide their thinking about which actions matter most and how they should be organized and successfully implemented. (Tinto, 2012, pp. 3–4)

The question that guided the current study was “Why do some students fail to experience the benefits of the enriching academic and social programs offered by colleges?” Theories and models addressing this question abound, but there is a growing consensus that floundering students tend to suffer deficits in the strength of their commitments to the school they are attending and to earning a degree (Tinto, 1993, pp. 41–44). In causal models, these two types of commitment are usually located directly before the end point outcomes of persistence and performance (Bean, 1985; Braxton, Duster, & Pascarella, 1988; Cabrera, Nora, & Castaneda, 1993). We consider these two types of commitment the pillars upon which students build their success.

What commitment means

The term “commitment” refers to a mind-set in which the person has strong, positive intentions toward a course of action or to an entity. Meyer and Allen (1991, 1997) developed a three-component model of commitment that includes positive affects, obligation to remain, and perceived costs of leaving. Translated into the setting of higher education, the three components describe the students’ level of commitment in terms of their positive or negative feelings about what they are doing and where they are doing it, their sense of dedication to keep doing it, and their estimate of the negative repercussions of abandoning it.

In higher education, commitment to a course of action means the pursuit of a degree, and commitment to an entity means attitudinal attachments to the school itself. Even though the two types of commitment are predictive of important student outcomes mentioned earlier, it is a complex puzzle as to why some students at a given school develop these qualities and others do not. Perhaps there are precursors to the development of commitment, individual differences among students which increase or decrease their ability to form commitments.

Davidson, Beck, and Grisaffe (2015) constructed and confirmed a nomological network of several psychological precursors, including several personality variables, one of which focused on our primary interest in the differences among students in their levels of motivation. They labeled it the “motivation to learn.” It built upon previous models of student success that highlighted a spectrum of motivational variables, such as achievement, control, intrinsic versus extrinsic regulation, and affiliation, as reviewed in the meta-analytic study by Robbins et al. (2004). One theory in particular unites these motives, or derivatives of them, into a single organizational scheme, the self-determination theory (SDT). This theory guided our study of the basic motivational forces that operate in college students and how best to construe their fulfillment or lack thereof.

Tenets of self-determination theory

SDT has been used to explain people’s reactions in a variety of contexts. Comprehensive presentations of SDT can be found in the seminal handbook edited by the theory’s cofounders (Deci & Ryan, 2002), a review article by Deci and Ryan (2000), and meta-analytic studies (Hagger & Chatzisarantis, 2016; Van den Broeck, Ferris, Chang, & Rosen, 2016). Our concern was higher education, so we extracted SDT’s four-stage sequence as it applies to college students: (a) the social and environmental factors provided in the school setting affect (b) the degree to which students basic needs are satisfied in that setting, which mediates (c) where they are positioned on a continuum of perceived internal, autonomous regulation versus external regulation, which causes (d) a wide variety of outcomes such as well-being, performance, commitment, and persistence. The third stage in the sequence, internal versus external regulation, is where SDT elaborates most extensively with subtheories of the mechanisms and processes by which students operate in a self-determined way. Much of the literature cited above summarizes the empirical links between the self-determined way of being and the fourth stage of outcomes.

Our primary interest was in the second stage of the SDT model. The deficits that people experience in need satisfaction disrupt and undermine all the possibilities and opportunities for growth that follow in subsequent stages. There are three basic needs proposed by SDT, considered to be innate, essential, and universal (Ryan & Deci, 2002, pp. 22–27). According to SDT, it is not their strength that is important, but it is the degree to which they are satisfied. We mention the “strength” versus “satisfaction” issue to clarify the distinction between “motives,” which vary in strength and signify what people “want” (as characterized in the aforementioned motives of achievement, affiliation, and control), and “needs,” which demand satisfaction and represent what people must have in order to attain well-being.

Three basic needs

The three basic needs in SDT are nCompetence, nRelatedness, and nAutonomy (the letter “n” abbreviates the word “need”). Their inclusion in SDT builds on past, extensive inquiries into the motives of achievement (for reviews, see Elliott & Dweck, 2005; Weiner, 2013) and affiliation (for review, see Baumeister & Leary, 1995), and to a lesser extent, independence (Murray, 1938). As conceptualized in SDT, nCompetence leads people to seek challenges that are optimal for their capabilities, striving to maintain and enhance their capacities; nRelatedness refers to feelings of belongingness and being connected to others in caring relationships, being in secure communion or unity with them; and nAutonomy refers to feelings of freedom and independence, being the origin and source of one’s own actions.

The satisfaction of these three needs can be viewed from a hierarchical perspective that has three levels of generality (Vallerand & Ratelle, 2002). At the top is “global” that encompasses the full range of one’s experiences. Directly below it is “contextual,” which includes focused activities that take place in a particular type of setting such as education, leisure, and work. And the bottom layer is “situational,” which refers to specific activities done at a specific time.

Our study of needs in college students was aimed at the contextual level. We believe that assessing needs at this level is appropriate for predicting outcomes that occur within this same hierarchical level. This logic is based on the principle of compatibility, which states that attitudes will be better predictors of behavior when measured at the same level of specificity (Ajzen & Fishbein, 2005). The compatibility strategy has been used by many others who have studied the connection between SDT variables and setting-specific outcomes that occur in sports, interpersonal relationships, leisure, gambling, politics, volunteering, education, work, and so on (for a review of these instruments, see Deci and Ryan, 2002, p. 46).

Linkage between basic needs and commitment

It is reasonable to suggest that deficits in the satisfaction of college students’ needs might prevent them from experiencing the two aforementioned outcomes that are important to educators, commitments to earning a degree and to the school. This possibility is consistent with the engagement model of commitment in a different context, the workplace, as proposed by Meyer, Gagne, and Parfyonova (2010) and Meyer (2014). Their model maps the three SDT needs onto entity (organizational) commitment, including its defining components. Some research supports key parts of the engagement model. For example, Gregurass and Diefendorff (2009) found that employees’ feelings of attachment to their employer were associated with how well the employees’ three SDT needs were met in the employment setting. This finding supports the premise that people develop positive feelings toward settings in which multiple needs are satisfied. We believe that the engagement model of commitment holds great promise for understanding why so many college students underperform or fail to graduate. To the extent that the linking processes in one context (employment) work the same in other contexts (educational), then perhaps they suffer from weak commitments because their basic needs are unsatisfied.

Hypotheses

In primary and secondary educational settings, students are known to experience a wide variety of positive outcomes when their teachers support their autonomy: high achievement and engagement in challenging tasks, perceived competence, positive emotions about the classroom and its interpersonal context, and perceptions of self-control (for review, see Reeve, 2002). In postsecondary educational settings, the satisfaction of SDT’s basic needs has been found to raise the students’ engagement (vigor, dedication, absorption) in their studies (Sulea, van Beek, Sarbescu, Virga, & Schaufeli, 2015).

When the aforementioned outcomes in educational settings are experienced by students, they are likely to develop positive attitudes about being in that setting. Therefore, their feelings of loyalty, belongingness, and commitment to being there should be instilled, Institutional Commitment. Also, if their needs are being satisfied as they pursue a goal, then their view of the goal as being personally worthwhile should be strengthened, Degree Commitment.

We hypothesized that each of the three needs would be positively related to both types of commitment, degree and institutional. Broken down into dyadic relationships, there were six specific hypotheses: each of the three needs will be positively correlated with degree and Institutional Commitment.

Measurement strategy

The hypotheses were contextualized (i.e. specific to the higher education setting), so it was best to utilize measures that are also contextualized. This strategy maximizes the likelihood of making accurate predictions of outcomes that occur in a particular setting by using predictors that function in the same setting. Therefore, we sought to find or develop valid, contextualized measures of the constructs. As mentioned earlier, commitment has been the focus of many previous studies in higher education, and one line of research in particular offered an instrument with outstanding psychometric properties, the College Persistence Questionnaire (CPQ; Davidson et al., 2015; Davidson, Beck, & Milligan, 2009).

However, the search for a contextualized measure of the three needs did not yield a suitable instrument. The often-used Academic Motivation Scale (Vallerand et al., 1992, 1993) focuses on self-regulation processes that occur after or in conjunction with the basic needs but does not have subscales for the needs themselves. Other instruments assess the three basic needs but do so at the global level (Heckert et al., 2000; Johnston, & Finney, 2010; Reeve & Sickenius, 1994). In fact, other researchers who have studied these three needs together in college students have had to use a generalized measure, or they pieced together contextualized measures of needs similar to SDT’s but from different sources and with different formats (see D. A. Guiffrida, Lynch, Wall, & Abel, 2013). Therefore, we decided to develop an instrument that would be not only be compatible with the question formats used in the CPQ but which would also contribute to future investigations of the three needs as they function in college students.

Being able to measure the three needs with a single instrument offers several advantages. The response biases inherent in self-report measures operate the same when the questions are derived from the same source and the response formats are the same across different sets of items. Comparisons across different needs present fewer interpretation complexities because measurement error is equalized. Also, there is convenience to users in not having to assemble questions from different sources that have different formats.

The utility of brief measures

In studying psychological variables, there is a trend toward using brief measures that emphasize high fidelity to the underlying construct rather than longer versions that feature broad bandwidth (e.g. Ashton, Jackson, Paunonen, & Rothstein, 1995; Burisch, 1984; Jenkins & Griffith, 2004; Paunonen & Ashton, 2013). The brevity offers several advantages if the somewhat diminished psychometric qualities associated with brief measures are acceptable—it reduces the participants’ time, effort, fatigue, and boredom and may thereby increase the quality of responses. Also, relatively short scales accommodate research designs that have many variables rather than a few, without overtaxing the attention span of participants. Some examples of brief measures of personality constructs using only one-to-four items per trait are the Single-Item Self Esteem Scale (Robins, Hendin, & Trzesniewski, 2001), the Five-Item Personality Inventory and the Ten-Item Personality Inventory, which assess Big-Five personality dimensions (Gosling, Renfro, & Swann, 2003; Romero, Villar, Gomez-Fraquela, & Lopez-Romero, 2012), and the Activity-Feeling States Scales, which measures four needs associated with intrinsic motivation (Reeve & Sickenius, 1994).

The benefits of brief measures are especially important to educational researchers. Students are inundated with assessment, so they may have low motivation to answer too many questions. From the vantage point of researchers, in order to build current models that include many variables, they must either divide the testing into multiple sessions or use brief measures of key constructs. Brief instruments may suffice unless there is a need to partition a construct into its composing facets and examine them separately.

Research plan

The purposes of the current study were (a) to develop a brief questionnaire that measured the three basic needs, with contextualized wording and acceptable psychometric qualities, and (b) if the first purpose was attained, then to test the hypotheses. If the predicted associations were confirmed, then it would also verify the concurrent validity of the newly developed measurement instrument.

Method

Participants and procedure

The sample included 1257 students at two universities who completed the questionnaire online during the last half of a semester. The participation was voluntary; one option from among several alternatives for earning extra credit in general education courses or advanced psychology courses. The participants were 72% females and 28% males. Their ethnicities were 2.5% Asian, 5.9% Black, 9.1% Hispanic, 78.7% White, and 3.7% Other. In terms of the parental education of the participants, 12% were first-generation college students (neither parent attended college). The academic classification of the participants was 43% freshmen, 26.1% sophomores, 16.9% juniors, and 13.6% seniors.

Instrument

We developed a question pool of 21 objective items to measure the three needs. The item content was guided by perusal of previously published definitions of the needs and their key components (Ryan & Deci, 2002, pp. 6–9) and their measurement at global or contextual levels (Gagne, 2003; D. Guiffrida, Gouveia, Wall, & Seward, 2008; Johnston & Finney, 2010; Reeve & Sickenius, 1994). The wording of the items was contextualized to fit the academic setting. We operationalized them as follows: nCompetence is satisfied to the extent that students believe that they can continue improving, produce quality products, handle difficult tasks well, and correctly anticipate what they need to know to do well; nRelatedness is satisfied when students believe that they have close friendships and a fulfilling social life, feel comfortable around classmates and have a sense of connectedness, and can easily find caring others who will listen to personal problems; nAutonomy is satisfied when students view their actions as freely chosen rather than coerced, performed without being told what to do, and done without trying to win the appreciation of others. The questions asked about all of the aforementioned operationalizations of the needs, so they covered a broad bandwidth of their respective need.

Each question was answered on a five-point Likert-type scale. The response-choice labels varied depending on the wording of the question. For example, if a question asked “How often … ” then the end points of the response continuum were very often and very seldom; and if a question asked “How satisfied … ” then the end points were very satisfied and very dissatisfied. The response choices were assigned scores on along a dimension of favorability that varied between +2 (favorable) to −2 (unfavorable).

The students’ commitment to earning a degree and feelings of loyalty to the school were measured with two scales from the CPQ—Version 2 (Davidson et al., 2015), the three-item Degree Commitment Scale (sample item: “After beginning college, students sometimes discover that a college degree is not quite as important to them as it once was. How strong is your intention to persist in your pursuit of the degree, here or elsewhere?”) and the four-item Institutional Commitment Scale (sample item: “How confident are you that this is the right college or university for you?”). Previous research has established the validity of both scales (Davidson et al., 2009, 2015), and in the current sample of participants, the two measures had acceptable internal reliability coefficients: Degree Commitment α = .77; Institutional Commitment α = .79.

Results

The analysis was divided into two phases. First, several steps were taken to reduce the item pool of questions intended to measure the three needs; the goal was to find a set of items that were the best representatives of the three needs. With brevity in mind, the intent was to settle on the best three or four items for each need and form them into three scales, yielding an overall set of 9–12 items. If the first phase achieved its goal, then the second phase proceeded to test the hypothesized relationships between the scores on the needs variables and the two commitment measures.

Component analysis to reduce the item pool and form scales

To reduce the item pool of the needs questions, the criterion of internal consistency was used. A series of principle component analyses (PCAs) were performed on the 21 question scores. The Kaiser–Meyer–Olkin (KMO) measure verified sampling adequacy for the analysis, KMO = .834, which is far above the minimum acceptable level of .50 (Field, 2009, p. 647). Bartlett’s test of sphericity χ²= 4868.641, p < .0001, indicated that the correlations between the items were sufficiently large for the PCA. The first solution yielded five components with eigenvalues higher than 1.00: 4.51, 2.38, 1.707, 1.30, and 1.06. They explained 52.18% of the variance.

A review of the components suggested that the latter two were trivial or ambiguous. The item loadings were insufficient: the fourth component had five items with loadings>.4 but all also loaded highly on multiple components; the fifth component had only one item that loaded >.4. Therefore, we focused the next analytical steps on the first three components.

PCA was performed utilizing a direct oblimin on three components. We requested the oblique rotation to look at the correlations among the components. The coefficients were less than .20, indicating that the components were minimally correlated (Tabachnick & Fidell, 2013, pp. 699–700). Consequently, we requested the varimax orthogonal rotation. This analysis yielded 11 items with a desirable pattern of loadings, high loading on one component (>.60) and low loadings on the other two components. Each component had at least three items with high loadings, so the components met the minimum criterion for reliably defining the underlying construct (Tabachnic & Fidell, 2013, p. 699).

Given the sparse number of items, it is important that they demonstrate sensitivity to differences in people. Therefore, we calculated frequency distributions on each of the 11. Two items had highly skewed distributions; they were dropped because they did not sufficiently discriminate between levels of the underlying construct.

The PCA with varimax rotation on the final set of nine items explained 63.35% of the variance, and each of the three components made substantial contributions (eigenvalue): Component 1 = 24.58% (2.18), Component 2 = 20.07% (1.80), and Component 3 = 18.70% (1.66). Based on the content of the items, we labeled the components nRelatedness, nCompetence, and nAutonomy, respectively.

Each component consisted of three items (rotated loadings are in parentheses): nRelatedness, satisfaction with number of close friends at school (.88), satisfaction with overall social life at school (.89), and sense of connectedness with others on campus (.74); nCompetence, satisfaction with rate of improvement in courses (.77), satisfaction with quality of academic work done (.76), and feeling adept at anticipating test questions beforehand (.74); nAutonomy, others at school telling self what to do too often (.79), having to take actions against own will at school (.75), and feeling unappreciated by others at school (.69).

Before forming scale scores and testing the hypotheses, the data were screened for outliers and cases where students answered fewer than two of the three questions that composed each of the three components. Based on the Mahalanobis distance calculation for outliers and the criteria of missing scores on two or more of the questions within a need component, 31 cases were not retained for further analyses, resulting in a total N of 1226.

Scores on the questions within each component were grouped together to form scale scores by calculating an average score across the three items. The means, standard deviations, internal reliability coefficients, and intercorrelations of the scales are presented in Table 1. The statistics in the table are reported separately for males and females because there were differences between the two sexes. We refer to the instrument as the Basic Psychological Needs Questionnaire: College Context.

Table 1.

Intercorrelations, means, standard deviations, and reliability coefficients of scales as a function of sex.

Measure	1	2	3	4	5	M	SD
1. Degree Commitment	1.00	.36	.26	.44	.26	1.53	.67
2. Institutional Commitment	.42	1.00	.35	.30	.29	1.24	.86
3. Need for relatedness	.24	.38	1.00	.36	.31	0.90	.82
4. Need for competence	.36	.30	.34	1.00	.22	0.87	.66
5. Need for autonomy	.20	.20	.28	.17	1.00	0.69	.81
M	1.75	1.41	0.85	0.79	0.89
SD	.53	.80	.87	.67	.80
α	.77	.79	.80	.65	.61

Note: Intercorrelations for males (n = 352) are presented above the diagonal, and intercorrelations for females (n = 905) are presented below the diagonal. Mean scale scores range from −2 to +2. The alpha coefficient is calculated for males and females combined. SD: standard deviation.

Testing the hypotheses

Before testing the hypotheses, we examined the bivariate relationship between the two types of commitment, Degree Commitment and Institutional Commitment, to verify that they were quantitatively as well as conceptually distinct from one another. The correlation coefficient was statistically significant: r = .40, p < .0001, indicating that the two variables did share some common variance (15.60%), but the size of the relationship supported the merits of analyzing them separately.

The hypothesized relationships between the need scales and the two types of student commitment, to earning a degree (Degree Commitment) and to the school they were attending (Institutional Commitment), were tested using sequential multiple regression, with the commitment scores as dependent variables and two blocks of independent variables—student background characteristics in the first block and scale scores on the three needs in the second block. The first block containing background variables was intended to serve as a control for the possibility that some unmeasured differences between the participants might be nested within these factors: sex (male or female), ethnicity (Hispanic, White; there were too few cases to analyze other ethnicities), first-generation college student, and academic classification (freshman, sophomore, junior, and senior).

To summarize the regression results, we report figures for (a) the first block of background characteristics, then (b) the next step of adding the second block of needs, then (c) the overall equation, and finally (d) the contribution of the individual variables (presented in Table 2).

Table 2.

Predictors of Degree Commitment and Institutional Commitment.

Variable	B	β	t	p	Semipartial r²	95% CI
Degree Commitment
Sex	−.22	−.17	6.34	.00	−.17	[−.29, −.16]
Ethnicity Hispanic	−.09	−.04	1.31	.19	−.03	[−.22, .04]
Ethnicity White	.11	.08	2.33	.02	.06	[.02, .20]
First generation	−.08	−.04	1.56	.12	−.04	[−.17, .02]
Academic classification	.03	.05	1.85	.06	.05	[.00, .05]
Need for relatedness	.06	.08	3.01	.00	.08	[.02, .10]
Need for competence	.27	.31	11.29	.00	.29	[.22, .32]
Need for autonomy	.09	.13	4.82	.00	.12	[.06, .13]
Institutional Commitment
Sex	−.18	−.10	−3.88	.00	−.10	[−.28, −.09]
Ethnicity Hispanic	.01	.00	0.09	.93	.00	[−.18, .19]
Ethnicity White	.21	.11	3.25	.00	.08	[.08, .34]
First generation	−.12	−.05	−1.82	.07	−.05	[−.26, .01]
Academic classification	.07	.09	3.30	.00	.08	[.03, .10]
Need for relatedness	.26	.27	9.48	.00	.24	[.20, .31]
Need for competence	.19	.16	5.75	.00	.15	[.13, .26]
Need for autonomy	.12	.12	4.28	.00	.11	[.06, .17]

Note: Classification is coded 1 = freshman, 2 = sophomore, 3 = junior, 4 = senior. B: unstandardized coefficient; β: standardized coefficient; CI: confidence interval of B.

Regression for Degree Commitment

The results for each step in the regression were statistically significant: Block 1 (student background characteristics), F(5, 1224) = 15.56, p < .0001; R = .24, R²= .06; Block 2 (student needs), ΔF(3, 1221) = 79.94, p < .0001; R = .46, R²= .21; ΔR²= .15; overall equation, F(8, 1226) = 41.58, p < .0001. With the addition of the second block of variables into the equation, the change in explained variance was substantial (15%).

Table 2 presents the regression coefficients for the individual variables. Among the background variables in the first block, sex and the White ethnicity were statistically significant. In the second block, all three needs were statistically significant. The direction of these relationships indicated that high scores in Degree Commitment were associated with females, Whites, and favorable scores on the needs. The relationships between Degree Commitment and each of the three needs confirmed the hypotheses. The coefficient for nCompetence was especially strong relative to the other predictors.

Regression for Institutional Commitment

The results for each step in the regression were statistically significant: Block 1 (student background characteristics), F(5, 1224) = 11.53, p < .0001; R = .21, R²= .05; Block 2 (student needs), ΔF(3, 1221) = 81.49, p < .0001; R = .45, R²= .20; ΔR²= .15; overall equation, F(8, 1226) = 39.19, p < .0001. With the addition of the second block of variables into the equation, the change in explained variance was reliably improved (15%).

Table 2 presents the regression coefficients for the individual variables. Among the background variables in the first block, the coefficients were statistically significant for sex, White ethnicity, and academic classification. In the second block, all three needs were statistically significant. The direction of these relationships indicated that high scores in Institutional Commitment were associated with females, Whites, advanced academic standing, and favorable scores on the needs. The relationships between Institutional Commitment and each of the three needs confirmed the hypotheses. The coefficient for nRelatedness was especially strong relative to the other predictors.

Discussion

The purposes of the study were achieved. Suitable measures of the primary constructs were found (the two commitment variables) or developed (the three needs), and six hypotheses about their relationships were tested and confirmed. The first phase of the analyses substantiated the development of a brief, contextualized measure of the three needs posited by SDT: nCompetence, nRelatedness, and nAutonomy. The component analyses reduced the initial pool of 21 items to 9, and each of the 9 finalists demonstrated high fidelity to one and only one component. The finalists were evenly divided among the three components, which corresponded to the three needs they were intended to measure, so scale scores were calculated as the average score among the relevant items.

The needs scales demonstrated reasonable internal reliability, given their brevity. Scales with fewer items tend to have lower internal reliability coefficients even if their item-to-scale coefficients are adequate (Cortina, 1993). The recommended minimal standard for internal reliability coefficients varies depending on a number of practical issues, such as the bandwidth (breadth) of the underlying construct, whether the items are intended to measure its breadth, and the number of items used (Simms & Watson, 2007, pp. 251–252).

We considered the needs to be broad, and we constructed an item pool addressing that breadth. The strategy was intended to avoid redundancy or focusing narrowly on a small part of the bandwidth. With this in mind, the minimal standard we set for the alpha coefficient was between .65 and .70 based on the recommendations of John and Soto (2007) for constructing personality scales with acceptable inter-item correlations. Two of the scales attained the standard, but the third, the nAutonomy Scale, fell slightly below it (.61). The internal reliability coefficient for nAutonomy was close enough to the standard to allow hypothesis testing, given the large sample size, which reduces the degree to which residual (error) variance in the scale underestimates its relationship with other variables.

The second phase of the analyses proceeded to use these scale scores to test the hypothesized relationships with commitment. This second phase not only tested hypotheses, but the results provided information about the concurrent validity of the needs measurements. That is, if the needs scales were associated in expected ways with other constructs then that would support the notion that each scale did measure its intended construct.

The results confirmed the hypothesized relationships. Regression analyses found that reasonable amounts of variance in commitment scores were explained by the three need predictors, after controlling for differences among the participants in their background characteristics. To gauge the strength of the overall findings, we compared the two blocks of predictor variables, first in terms of the variance explained by the overall block and then the individual variables within the block.

The block that contained student backgrounds explained substantially and significantly less variance compared with the block that contained the three needs. Assuming that the three needs were based largely on students’ experiences, this finding highlights the importance of considering the experiences of students when trying to understand the processes by which they form commitments. This is not to say that backgrounds were irrelevant, because some of the background variables did matter, and the results can be helpful in identifying groups who might be at risk for weak commitments.

For example, men scored lower than women in both types of commitment. These findings might partially explain why men tend to have lower academic achievements than women and are also less likely to persist (Astin & Oseguera, 2012; Buchmann & DiPrete, 2006) because commitment, by definition, stimulates mental engagement in relevant tasks and a reluctance to abandon them. The sex differences in commitment found in the current study can be added to other differences that previous research has identified, such as family, personality, and social-risk characteristics (Pidcock, Fischer & Munsch, 2001).

Another background factor that was significantly associated with commitment was the academic classification variable. Lower level students tended to score lower than did advanced students. The relationship was statistically significant for Institutional Commitment but slightly less than significant for Degree Commitment (p = .06). The findings coincide with changes in persistence rates, which rise in the second and third years of four-year programs (Nora & Crisp, 2012) and can be partially explained by the previously mentioned, direct effect of commitment on persistence. Weakly committed students tend to drop out prematurely, so they were not equally represented in the sample of advanced students. Another explanation utilizes a developmental perspective. According to this viewpoint, the students’ commitment grows as a consequence of longevity and additional experiences at their college. Its growth co-occurs with a multitude of other positive progression processes, as summarized in Pascarella and Terenzini (2005), many of which might exert bidirectional forces on each other.

Overall, the analysis of the background factors is helpful in understanding the vulnerability of specific groups of students whose weak commitment(s) might produce undesirable outcomes. The sample of convenience in this study was diverse enough to include these background characteristics as control variables, but future research with larger subgroups is needed to explore interactions with the needs in the second block, which might clarify some of the reasons for explaining high or low commitment scores.

While the analysis of the background variables yielded some insights into the students’ commitment, inclusion of the basic needs variables into the regression equations tripled the amount of variance explained in commitment scores, resulting in R² of .21 (degree) and .20 (institutional). Given the multidetermined nature of commitment and the wide range of student experiences that influence it (for recent causal model, see Davidson et al., 2015), we interpret the needs variables, taken as a group, to be salient and worthwhile including in future model development to test their fit alongside other established factors.

After considering the importance of three needs as a group (or block), it is necessary to evaluate them individually, as per the hypotheses. The results of the regression equations did confirm the hypotheses, yielding six statistically significant coefficients for the six expected dyadic relationships. The regression technique partialled out overlapping variance among the needs, so their relationships with the commitment variables were autonomous and genuine, which simplified the interpretation of the relationships.

Among the six, two stood out from others in their strength: nCompetence with Degree Commitment and nRelatedness with Institutional Commitment. Clearly, self-appraisals of competence are important in students as they pursue the academic tasks and courses that eventuate in a degree, and experiencing positive feelings about their social contacts helps instill thoughts that the school is the right place to be.

Undoubtedly, there are many ways to construe the meaning of these two relationships, and future research can enlighten the issue. In the meantime, we will mention some possibilities. In the case of nCompetence, deficits may cause students to seek unchallenging, easy tasks and courses in order to establish minimal levels. Such a strategy might distract or delay or derail them from envisioning themselves as a degree-capable person and pursuing challenging degree requirements. This interpretation is consistent with voluminous research on the role of self-efficacy in higher education, as summarized in the meta-analytic research of Multon, Brown, and Lent (1991). They reported that low-efficacious students, in comparison to high-efficacy students, tended to attempt less (in terms of the number of academic tasks undertaken) and accomplish less (in terms of the number of academic tasks completed, and in terms of grade point average).

Another possible reason for the weak Degree Commitment among students with low-satisfaction levels of nCompetence is their tainted perception of the goal. That is, when people perceive a desirable goal as unfeasible, it lessens the personal importance they place on the goal, and it also weakens their energization to pursue it (Oettingen et al., 2009).

With regard to the connection between nRelatedness and Institutional Commitment, students might not be able to form strong positive loyalties to a school without first establishing affirming social relationships there, which would be consistent with many previously verified views of the social integration of college students (Braxton & McClendon, 2001; Stoecker, Pascarella, & Wolfle, 1988; Tinto, 1993; Vianden & Barlow, 2014). Students with deficits in nRelatedness may be tempted to pursue unhealthy relationships that distract them from fully embracing the school and its prevailing value systems. For example, they might overly pursue relationships with others who offer acceptance and approval but have very little to offer in terms of enriching the students’ personal, academic, and career goals.

While nCompetence and nRelatedness were much more strongly related to one type of commitment as opposed to the other type, nAutonomy was similarly related to both types of commitment. This may mean that feelings of freedom and volition are equally fundamental to forming intentions to persist along the path toward a goal (Degree Commitment) as well as developing feelings of loyalty toward those who are coparticipants in the path (Institutional Commitment). This interpretation is consistent with SDT’s view of the processes that produce internalization, transforming external requests and expectations into personally endorsed values and self-regulation: nCompetence is especially important in internalizing the value of achievement activities, nRelatedness is especially crucial in internalizing the merits of social activities, and nAutonomy is essential to both (Deci & Moller, 2005, pp. 589–591).

Limitations

The research design was correlational, so statements about cause and effect must be tempered by acknowledging that the results do not substantiate causal relationships. The theoretical perspective that guided this study, SDT, proposes that the satisfaction of basic needs does exert causal force on outcomes such as peoples’ commitment and determination to persist in pursuit of a goal and sustain their belief in their goodness of fit within a setting. The findings of this study are consistent with this theoretical proposition.

Ordinarily, the next step in scale development would be to perform a confirmatory factor analysis (CFA). Given the broad nature of the needs, in terms of their definitions and operationalizations, and given the small number of items to used measure them, we believe it is premature to subject scores to the rigorous standards of CFA. Even the best personality inventories often fail to fit CFA models because of the trade-off between (a) the fidelity of the items to the underlying trait or motive and (b) the degree to which the items address the bandwidth (Hopwood & Donnellan, 2010).

The PCA results in the current study are encouraging and can serve as a foundation for future research. The schools and participants were not drawn randomly, so questions of generality emerge. To determine the external validity of our findings, we propose that the Basic Psychological Needs Questionnaire: College Context be administered to undergraduates with different background characteristics than those in the current study. Efforts should also be made to assess students from other institutions and types of programs (e.g. small private schools, online, etc.).

Even though the reliability coefficients approximated the minimal standard we set, the residual error in the scale scores undoubtedly reduced the relationships with the two commitment variables. Administering the instrument to a group of students on multiple occasions would contribute another indicator of the instrument’s reliability.

Like any questionnaire, the Basic Psychological Needs Questionnaire: College Context could be improved by the addition of new items and/or the replacement of current items. Therefore, we suggest that future tests of the instrument include several new questions whose contributions could be assessed via a PCA. Increases in the alpha coefficient of the nAutonomy Scale would be particularly desirable. If these results are promising, a second sample could be obtained, a CFA and a subsequent structural equation model tested. While several theoretical models could be proposed, our current thinking, and the results of this investigation, suggest that each of the three needs is directly associated with Institutional Commitment and Degree Commitment.

Including a few additional items might improve each scale’s bandwidth and raise the internal reliability coefficients. Also, the instrument’s predictive validity can be expanded by testing relationships with a wide range of outcomes that are important to students and school personnel such as academic accomplishments, participation in activities in and out of the classroom, degree attainment, joining in alumni affairs and networks, and postgraduate employment success.

Implications

The Basic Psychological Needs Questionnaire: College Context can be used either as a stand-alone instrument or the items can serve as a core upon which to build additional indicators of the three needs. Our goal was to produce a brief device, but certainly there is room for additional items. It is worth noting that in appraising the reliability of brief measures, perhaps the stability of scores across time (test–retest) is as important as across items (internal). The internal reliability coefficients were acceptable for our purposes but somewhat lower than lengthier measures. Future research should examine the test–retest reliability of the nine items.

The assessment of needs is important to those who build models for understanding the crucial factors in college students’ commitment, performance, and retention. The inclusion of the three needs we assessed is important not only to SDT researchers but also to others who seek to include motivational factors in their models. The results of this study indicate that motivational factors should be included in comprehensive models of student outcomes, especially nCompetence, nRelatedness, and nAutonomy. The instrument we developed can aid their work by raising the predictability of important outcome variables not only because of the significance of the three needs but also because the measurements are contextualized, that is, the wording of the items is grounded in the same context as the outcome variables. The brevity of the instrument makes it easier for comprehensive models to test the veracity of a wide range of potentially relevant factors.

The personnel of institutions of higher education are concerned with identifying at-risk students and developing effective interventions. The three need scales developed in this study offer an effective tool with which to both identify such students and guide the process of designing interventions to correct deficits. We recommend that the nine-item measurement of needs be combined with the seven CPQ items that measure commitment, forming a 16-question survey.

Unfavorable scores on the survey will indicate which students need help and also point to specific areas of their vulnerabilities. The deficits that are identified on such a questionnaire speak to the issue of pinpointing sources of problems that surface in a wide variety of ways, most of which eventually come to the attention of college personnel. Such assessments are often included in freshmen orientation courses. The brevity of the instrument enables students to complete it quickly, so it can be incorporated rather easily into course assignments without much cost in terms of students’ time.

From a programmatic perspective, schools might use the Basic Psychological Needs Questionnaire: College Context to appraise the impact of initiatives that are supposed to raise students’ academic capabilities such as tutoring (nCompetence), extracurricular student life activities (nRelatedness), or workshops on options in degrees, majors, concentrations, and career opportunities (nAutonomy).

The path that college students travel to finish their training and attain a degree is lengthy, difficult, and filled with obstacles and challenges that must be dealt with. The development of commitment is essential to being successful in the higher educational environment. The results of this study offer some ideas about the psychological factors than can help or hinder their capabilities to build strong commitments to what they are doing and to where they are doing it.

Footnotes

Article Notes

References

American College Testing (ACT) (2010). What works in student retention? Fourth national survey. Public four-year colleges and universities report. Retrieved from https://www.google.com/?gws_rd=ssl#q=ACT+fourth+national+survey.

Ajzen

Fishbein

(2005) The influence of attitudes on behavior. In: Albarracín

Johnson

B. T.

Zanna

M. P.

(eds) The handbook of attitudes, Mahwah, NJ: Lawrence Erlbaum Associates, pp. 173–221.

Ashton

M. C.

Jackson

D. N.

Paunonen

S. V.

Rothstein

M. G.

(1995) The criterion validity of broad factor scales versus specific trait scales. Journal of Research in Personality 29(4): 432–442. doi:10.1006/jrpe.1995.1025.

Astin

A. W.

Oseguera

(2012) Pre-college and institutional influences on degree attainment. In: Seidman

(ed.) College student retention, formula for student success, 2nd ed. New York, NY: Rowman & Littlefield, pp. 119–146.

Baumeister

Leary

M. R.

(1995) The need to belong: Desire for interpersonal attachments as fundamental human motivation. Psychological Bulletin 117(3): 497–529. doi:10.1037/0033-2909.117.3.497.

Bean

J. P.

(1985) Interaction effects based on class level in an explanatory model of college student dropout syndrome. American Educational Research Journal 22(1): 35–64.

Braxton

J. M.

Duster

Pascarella

E. T.

(1988) Causal modeling and path analysis: An introduction and an illustration in student attrition research. Journal of College Student Development 29: 263–272.

Braxton

J. M.

McClendon

S. A.

(2001) The fostering of social integration and retention through institutional practice. Journal of College Student Retention: Research, Theory & Practice 3(1): 57–71. doi:10.2190/RGXJ-U08C-06VB-JK7D.

Buchmann

DiPrete

T. A.

(2006) The growing female advantage in college completion: The role of parental resources and academic achievement. American Sociological Review 71: 515–541.

10.

Burisch

(1984) You don’t always get what you pay for: Measuring depression with short and simple versus long and complicated scales. Journal of Research in Personality 18(1): 81–98. doi:10.1016/0092-6566(84)90040-0.

11.

Cabrera

A. F.

Nora

Castaneda

M. B.

(1993) College persistence: Structural equations modeling test of an integrated model of student retention. Journal of Research in Higher Education 64(2): 123–137.

12.

Cortina

J. M.

(1993) What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology 78(1): 98–104.

13.

Davidson

W. B.

Beck

H. P.

Grisaffe

D. B.

(2015) Increasing the institutional commitment of college students: Enhanced measurement and test of a nomological model. Journal of College Student Retention: Research, Theory & Practice 17(2): 162–185. doi:10.1177/1521025115578230.

14.

Davidson

W. B.

Beck

H. P.

Milligan

(2009) The College Persistence Questionnaire: Development and validation of an instrument that predicts student attrition. Journal of College Student Development 50(4): 373–390. doi:10.1353/csd.0.0079.

15.

Deci

E. L.

Moller

A. C.

(2005) The concept of competence: A starting place for understanding intrinsic motivation and self-determined extrinsic motivation. In: Elliott

A. J.

Dweck

C. S.

(eds) Handbook of competence and motivation, New York, NY: Guilford Press, pp. 579–597.

16.

Deci

E. L.

Ryan

R. M.

(2000) The “what” and “why” of goal pursuits: Human needs and the self-determination of behavior. Psychological Inquiry 11(4): 227–268. doi:10.1207/S15327965PLI1104_01.

17.

Deci

E. L.

Ryan

R. M.

(2002) Handbook of self-determination research, Rochester, NY: University of Rochester Press.

18.

Elliott

A. J.

Dweck

C. S.

(2005) Handbook of competence and motivation, New York, NY: Guilford Press.

19.

Field

(2009) Discovering statistics using SPSS, 3rd ed. Thousand Oaks, CA: SAGE.

20.

Gagne

(2003) The role of autonomy support and autonomy orientation in prosocial behavioral engagement. Motivation and Emotion 27(3): 161–186.

21.

Gosling

S. D.

Renfro

P. J.

Swann

W. B.

(2003) A very brief measure of the Big-Five personality domains. Journal of Research in Personality 37(6): 504–528. doi:10.1016/S0092-6566(03)00046-1.

22.

Greguras

G. J.

Diefendorff

J. M.

(2009) Different fits satisfy different needs: Linking person-environment fit to employee commitment and performance using self-determination theory. Journal of Applied Psychology 94(2): 465–477. doi:10.1037/a0014068.

23.

Guiffrida

Gouveia

Wall

Seward

(2008) Development and validation of the Need for Relatedness at College Questionnaire (NRC-Q). Journal of Diversity in Higher Education 1(4): 251–261. doi:10.1037/a0014051.

24.

Guiffrida

D. A.

Lynch

M. F.

Wall

A. F.

Abel

D. S.

(2013) Do reasons for attending college affect academic outcomes? A test of a motivational model from a self-determination theory perspective. Journal of College Student Development 54(2): 121–139. doi:10.1353/csd.2013.0019.

25.

Hagger

M. S.

Chatzisarantis

N. L. D.

(2016) The trans-contextual model of autonomous motivation in education: Conceptual and empirical issues and meta-analysis. Review of Educational Research 86(2): 360–407. doi:10.3102/0034654315585005.

26.

Heckert

T. M.

Cuneio

Hannah

A. P.

Adams

P. J.

Droste

H. E.

Mueller

M. A.

Roberts

L. L.

(2000) Creation of a new Needs Assessment Questionnaire. Journal of Social Behavior & Personality 15(1): 121–136.

27.

Hopwood

C. J.

Donnellan

M. B.

(2010) How should the internal structure of personality inventories be evaluated? Personality and Social Psychology Review 14(3): 332–346. doi:10.1177/088868310361240.

28.

Jenkins

Griffith

(2004) Using personality constructs to predict performance: Narrow or broad bandwidth. Journal of Business and Psychology 12(2): 255–269. doi:10.1037/t36519-000.

29.

John

O. P.

Soto

C. J.

(2007) The importance of being valid: Reliability and the process of construct validation. In: Robins

R. W.

Fraley

R. C.

Krueger

R. F.

(eds) Handbook of research methods in personality psychology, New York, NY: Guilford Press, pp. 461–494.

30.

Johnston

M. M.

Finney

S. J.

(2010) Measuring basic needs satisfaction: Evaluating previous research and conducting new psychometric evaluations of the Basic Needs Satisfaction in General Scale. Contemporary Educational Psychology 35(4): 280–296. doi:10.1016/j.cedpsych.2010.04.003.

31.

Meyer

J. P.

(2014) Employee commitment, motivation, and engagement: Explaining the links. In: Gagne

(ed.) The Oxford handbook of work engagement, motivation, and self-determination theory, New York, NY: Oxford University Press, pp. 33–49.

32.

Meyer

J. P.

Allen

N. J.

(1991) A three-component conceptualization of organizational commitment. Human Resources Management Review 1: 64–89.

33.

Meyer

J. P.

Allen

N. J.

(1997) Commitment in the workplace: Theory, research, and application, Thousand Oaks, CA: SAGE.

34.

Meyer

J. P.

Gagne

Parfyonova

N. M.

(2010) Toward an evidence-based model of engagement: What we can learn from motivation and commitment research. In: Albrecht

(ed.) The handbook of employee engagement: Perspectives, issues, research, and practice, Chelrenham, England: Edward Elgar Publishing, pp. 62–73.

35.

Multon

K. D.

Brown

S. D.

Lent

R. W.

(1991) Relation of self-efficacy beliefs to academic outcomes: A meta-analytic investigation. Journal of Counseling Psychology 38(1): 30–38. doi:10.1037/0022-0167.38.1.30.

36.

Murray

H. A.

(1938) Explorations in personality: A clinical and experimental study of fifty men of college age, New York, NY: Oxford University Press.

37.

National Center for Education Statistics (2016). Digest education statistics, 2015— Chapter 3: Postsecondary education. Retrieved from National Center for Education Statistics website: https://nces.ed.gov/programs/digest/d15/index.asp.

38.

Nora

Crisp

(2012) Student persistence and degree attainment beyond the first year in college. In: Seidman

(ed.) College student retention, formula for student success, 2nd ed. New York, NY: Rowman & Littlefield, pp. 229–249.

39.

Oettingen

Mayer

Sevincer

A. T.

Stephens

E. J.

Pak

Hagenah

(2009) Mental contrasting and goal commitment: The mediating role of energization. Personality and Social Psychology Bulletin 35(5): 608–622. doi:10.1177/0146167208330856.

40.

Pascarella

E. T.

Terenzini

P. T.

(2005) How college affects students: A third decade of research Vol. 2, San Francisco, CA: Jossey-Bass.

41.

Paunonen

S. V.

Ashton

M. C.

(2013) On the prediction of academic performance with personality traits: A replication study. Journal of Research in Personality 47(6): 778–781. doi:10.1037/t18846-000.

42.

Pidcock

B. W.

Fischer

J. L.

Munsch

(2001) Family, personality, and social risk factors impacting the retention rates of first-year Hispanic and Anglo college students. Adolescence 36(144): 803–818.

43.

Reeve

(2002) Self-determination theory applied to educational settings. In: Deci

Ryan

(eds) Handbook of self-determination research, Rochester, NY: University of Rochester Press, pp. 182–203.

44.

Reeve

Sickenius

(1994) Development and validation of a brief measure of the three psychological needs underlying motivation: The AFS scales. Educational and Psychological Measurement 54(2): 506–515. doi:10.1177/0013164494054002025.

45.

Robbins

S. B.

Lauver

Davidso

Langley

(2004) Do psychosocial and study skill factors predict college outcomes? A meta-analysis. Psychological Bulletin 130(2): 261–288. doi:10.1037/0033-2909.130.2.261.

46.

Robins

R. W.

Hendin

H. M.

Trzesniewski

K. H.

(2001) Measuring global self-esteem construct validation of a single-item measure of the Rosenberg Self-Esteem scale. Personality and Social Psychology Bulletin 27: 151–161. doi:10.1177/0146167201272002.

47.

Romero

Villar

Gomez-Fraguela

J. A.

Lopez-Romero

(2012) Measuring personality traits with ultra-short scales: A study of the Ten Item Personality Inventory (TIPI) in a Spanish sample. Personality and Individual Differences 53: 289–293. doi:10.1016/j.paid2012.03.035.

48.

Ryan

R. M.

Deci

E. L.

(2002) Overview of self-determination theory: An organismic dialectic perspective. In: Deci

Ryan

(eds) Handbook of self-determination research, Rochester, NY: University of Rochester Press, pp. 6–9, 22–27.

49.

Shapiro, D., Dundar, A., Wakhungu, P. K., Yuan, X., Nathan, A., & Hwang, Y. (2015, November). Completing college: A national view of student attainment rates—Fall 2009 cohort (Signature Report No. 10). Herndon, VA: National Student Clearinghouse Research Center.

50.

Simms

L. J.

Watson

(2007) The construct validation approach to personality scale construction. In: Robins

R. W.

Fraley

C. R.

Krueger

R. F.

(eds) Handbook of research methods in personality psychology, New York, NY: Guilford Press, pp. 240–258.

51.

Stoecker

Pascarella

E. T.

Wolfle

L. M.

(1988) Persistence in higher education: A 9-year test of a theoretical model. Journal of College Student Development 29: 198–209.

52.

Sulea

van Beek

Sarbescu

Virga

Schaufeli

W. B.

(2015) Engagement, boredom, and burnout among students: Basic need satisfaction matters more than personality traits. Learning and Individual Differences 42: 132–138. doi:10.1016/j.lindif.2015.08.018.

53.

Tabachnick

B. G.

Fidell

L. S.

(2013) Using multivariate statistics, Noida, India: Pearson Education.

54.

Tinto

(1993) Leaving college: Rethinking the causes and cures of student attrition, Chicago, IL: University of Chicago Press.

55.

Tinto

(2012) Completing college: Rethinking institutional action, Chicago, IL: University of Chicago Press.

56.

Vallerand

R. J.

Pelletier

Blais

M. R.

Briere

N. M.

Senecal

Vallieres

E. F.

(1992) The Academic Motivation Scale: A measure of intrinsic, extrinsic, and amotivation in education. Educational and Psychological Measurement 52(4): 1003–1017. doi:10.1177/0013164492052004025.

57.

Vallerand

R. J.

Pelletier

Blais

M. R.

Briere

N. M.

Senecal

Vallieres

E. F.

(1993) On the assessment of intrinsic, extrinsic, and amotivation in education: Evidence on the concurrent and construct validity of the Academic Motivation Scale. Educational and Psychological Measurement 53(1): 159–172. doi:10.1177/0013164493053001010.

58.

Vallerand

R. J.

Ratelle

C. F.

(2002) Intrinsic and extrinsic motivation: A hierarchical model. In: Deci

Ryan

(eds) Handbook of self-determination research, Rochester, NY: University of Rochester Press, pp. 37–63.

59.

Van den Broeck

Ferris

D. L.

Chang

Rosen

C. C.

(2016) A review of self-determination theory’s basic psychological needs at work. Journal of Management 42(5): 1195–1229. doi:10.1177/0149206316632058.

60.

Vianden

Barlow

P. J.

(2014) Showing the love: Predictors of student loyalty to undergraduate institutions. Journal of Student Affairs Research 51(1): 16–29. doi:10.1515/jsarp-2014-002.

61.

Weiner

(2013) Little-known truths, quirky anecdotes, seething scandals, and even some science in the history of (primarily achievement) motivation. Personality and Social Psychology Review 17(3): 293–304. doi:10.1177/1088868313484587.