Self-Regulatory Climate: A Social Resource for Student Regulation and Achievement
by Curt M. Adams, Patrick B. Forsyth , Ellen Dollarhide , Ryan Miskell & Jordan Ware - 2015
Background/Context: Schools have differential effects on student learning and development, but research has not generated much explanatory evidence of the social-psychological pathway to better achievement outcomes. Explanatory evidence of how normative conditions enable students to thrive is particularly relevant in the urban context where attention disproportionately centers on the pathology of these environments rather than social attributes that contribute to student growth.
Research Purpose: Our purpose in this study was to determine if a self-regulatory climate works through student self-regulation to influence academic achievement. We hypothesized that (1) self-regulatory climate explains school-level differences in self-regulated learning, and (2) self-regulated learning mediates the relationship between self-regulatory climate and math achievement.
Research Design: We used ex post facto survey data from students and teachers in 80 elementary and secondary schools from a large, southwestern urban school district. A multilevel modeling building process in HLM 7.0 was used to test our hypotheses.
Results: Both hypotheses were supported. Self-regulatory climate explained significant school-level variance in self-regulated learning. Additionally, student self-regulated learning mediated the relationship between self-regulatory climate and math achievement.
Conclusions: Our results suggest that schools, like teachers, have differential effects on the motivational resources of students, with self-regulatory climate being an essential social condition for self-regulation and achievement. We believe self-regulatory climate has value for educators seeking to provide equitable learning opportunities for all students and for researchers seeking to account for achievement differences attributed to schools. In both cases, self-regulatory climate advances a construct and measure that conceptualizes and operationalizes school-level support for psychological needs.
Research on student motivation and behavior indicates that student interest in learning, engagement in academic tasks, and achievement partly depend on teachers who support student autonomy, competence, and relatedness (Reeve, 2006). Conversely, teachers who control student behavior and regulate performance largely with external contingencies like threats or rewards can end up undermining student interest, engagement, and motivation (Vansteenskiste, Simons, Lens, Soenens, & Matos, 2005). If student motivation to perform academic tasks depends in part on teachers and classroom environments (Reeve, 2002, 2006), it seems reasonable to believe that the social organization of schools can also nurture or thwart the capacity of students to thrive academically.
Indeed, research on school climate and culture has identified conditions in schools that enhance teaching and facilitate deep learning (Cohen McCabe, Michelli, & Pickeral, 2009; Hoy, 2012). We know, for instance, that teachers and students perform better in schools where collective trust (Adams & Forsyth, 2013; Goddard, Salloum, & Berebitsky, 2009), academic optimism (Hoy, Tarter, & Woolfolk Hoy, 2006), and social capital (Leana & Pil, 2006) exist. This research, however, has not established an explanatory link connecting the social organization of schools to student outcomes. Evidence of how organizational conditions enable students to thrive is particularly relevant in the urban context where attention disproportionately centers on the pathology of these environments rather than on social attributes that contribute to student growth. Our purpose in this study was to determine if a self-regulatory climate works through student self-regulation to influence academic achievement.
CONCEPTUALIZATION OF SELF-REGULATORY CLIMATE
Self-regulated students are metacognitvely, motivationally, and behaviorally active learners (Zimmerman, 1990). They act volitionally toward academic goals and possess the inner agency to control academic efforts (Reeve, Ryan, Deci, & Jang, 2008). Self-regulatory climate is not an individual state, nor is it the sum of individual regulatory beliefs of students in a school. Instead, it is a product of social features of a school that bring to life cooperative, trustworthy, and academically focused interactions. We define a self-regulatory climate as a set of normative conditions formed through studentteacher interactions that are capable of satisfying student psychological needs (Adams, Ware, Miskell, Dollarhide, & Forsyth, in press).
What normative conditions would be capable of satisfying student psychological needs? We turn to self-determination theory to answer this question. Ryan and Deci (2000) define a need as an energizing state that when satisfied fuels growth and well-being, but when neglected leads to underperformance and distress. Support for psychological needs comes from an external context where healthy, cooperative relationships enable the internalization of learning (Deci & Ryan, 2000, 2002). Autonomy support, competence support, and relational support characterize task contexts that leverage student capacity as a vital resource for motivated academic behavior (Reeve, 2006).
SCHOOL SUPPORT FOR STUDENT PSYCHOLOGICAL NEEDS
Autonomy support reflects cooperative and trustworthy interactions between teachers and students that enable teachers to create a student-centered environment (Assor, Kaplan, & Roth, 2002). Schools support student autonomy by emphasizing relevance in learning, relying on non-controlling informational language and structures to engage students, encouraging choice in the selection of tasks and projects, and allowing for independent thinking (Deci, Eghrari, Patrick, & Leone, 1994; Jang, Reeve, & Deci, 2010; Reeve & Jang, 2006). Competence support reflects a school context that promotes agentive beliefs and volitional behavior by setting students up to experience academic success rather than failure (Niemiec & Ryan, 2009; Reeve, 2006). Relational support emerges through generative studentteacher interactions and attachments that promote feelings of belonging and security (Reeve et al., 2008; Ryan & Deci, 2000). We believe school-level support for psychological needs exists in a climate where collective trust and academic emphasis enable and fosters internal regulation.
Collective faculty trust in students has implications for how autonomy supportive or autonomy suppressive the school climate feels to students. Collective faculty trust signals an established pattern of responsible and trustworthy student behavior that enables teachers to regulate learning through relationships and social influence (Forsyth, Adams, & Hoy, 2011). The presence of faculty trust obviates the need to use strong, external controls that force student compliance with rules, regulations, and expectations (Forsyth et al., 2011). A lack of trust signals a climate where students cannot be counted on to control their behavior without some identified contingency. With low trust, autonomy-suppressive practices like restricting student criticism and independent opinions, forcing compliance with meaningless and uninteresting activities, or using external inducements to encourage expected behavior become common practices (Assor, Kaplan, & Roth, 2002). It is hard to envision autonomy-supportive structures and practices in the absence of collective faculty trust in students.
When faculty trust in students is reciprocated with high levels of collective student trust in teachers, a context is in place that enables relational support at the school level. Collective student trust signals a relational environment where students are connected to teachers and stimulated by instructional activities. Low student trust signals relational tension that can thwart internal motivation and authentic engagement. Empirically, student trust has positive motivational and behavioral consequences. Ryan, Stiller, and Lynch (1994) found that students experienced greater intrinsic motivation when they perceived teachers as caring and honest. Connell and Wellborn (1991) found that students emotional security with classmates and teachers predicted patterns of action that in turn predicted academic performance. Trust in general provokes an inner determination to work toward goals, to take responsibility for actions, and to engage authentically in tasks (Darley, 2004). Student trust in teachers encourages the type of risk taking that is necessary for students to put forth quality effort in school and to become autonomous learners (Ryan & Deci, 2002).
Collective trust operates very differently than external control. Trust establishes cooperation and predictability through discretion, choice, and risk taking. External control, in contrast, creates predictability by constraining behavior and forcing compliance with rules and expectations (Adams, Ware, Miskell, Dollarhide, & Forsyth, in press). The evidence suggests that mutual facultystudent trust increases the likelihood that teachers will choose to work toward a common vision, take shared responsibility for learning and achievement, promote discretion and choice, foster independent thinking and problem solving, and maintain high expectations for student success (Forsyth et al., 2011). Collective trust, as a form of social control, works through the internal capacity of teachers and students to drive beliefs and behaviors that sustain school engagement.
The second constitutive dimension of a self-regulatory climate emerges from student-perceived academic emphasis. Academic emphasis is conceptually defined as the degree to which schools set high academic expectations and work collectively toward achieving academic excellence (Goddard, Sweetland, & Hoy, 2000; Hoy & Sabo, 1998). Processes and practices that embody high academic expectations and emphasis include setting achievable goals, maintaining a coherent and orderly learning environment, pressing students to work hard in class, and celebrating academic excellence (Hoy et al., 2006). Competence, as understood from self-determination theory, is supported when students experience teacher encouragement, consistent high expectations, and emphasis on understanding instructional concepts (Reeve, 2002, 2006). When a schools high level of academic emphasis is palpable, it is part of a climate that we call self-regulatory because it relies less on external sanctions and urges toward the formation of internally motivated student behavior.
SELF-REGULATORY CLIMATE: A DISTINCT CONCEPT
Figure 1 shows the three observable features that together constitute a school climate capable of nurturing internal student regulation and quality school performance. The presence of only one trust form, or simply having academic emphasis without trust, does not constitute a self-regulatory climate. All three conditions are interrelated and necessary to form a climate that enhances the internal capacity of students. For instance, faculty trust emerges when teachers experience students as being interested in academic tasks and engaged in learning activities. Student trust becomes normative when students experience teachers as genuinely committed to their learning and well-being. Academic emphasis is observable in teacher and student behaviors that are consistent with high academic achievemen (Adams, Ware, Miskell, Dollarhide, & Forsyth, in press). The two dimensions constituting a self-regulating climate gauge the experiences of both teachers and students, reflecting the beliefs and behavior of the two role groups most directly responsible for the learning climate in the school.
Figure 1. The interdependent properties of self-regulatory climate. SRC Self-Regulatory Climate; Student Trust in Teachers; Faculty Trust in Students; Student-Perceived Academic Emphasis.
In a previous study we found that collective faculty trust in students, collective student trust in teachers, and academic emphasis were indeed interrelated normative conditions (Adams, Ware, Miskell, Dollarhide, & Forsyth, in press). These conditions coalesce in schools to define a climate produced by healthy, cooperative, and trustworthy relationships where students and teachers are perceived as striving toward academic success. Each observed factor of the self-regulatory climate is capable of nurturing student psychological needs. We also found a strong relationship between self-regulatory climate and school performance after controlling for the influence of prior achievement, school socio-economic status, and minority composition. Evidence from our initial exploration suggests an association between self-regulatory climate and school performance, but it did not address the social-psychological pathway by which a self-regulatory climate may contribute to student achievement.
RATIONALE AND HYPOTHESES
We turn to evidence based in self-determination theory to understand the effect of a self-regulatory climate on student beliefs, behavior, and achievement. Self-determination theory proposes that social environments supportive of autonomy, competence, and relatedness advance learning by using the natural internal capacity of students as a resource to trigger self-regulated and self-determined behavior (Deci & Ryan, 1985, 2008; Ryan & Deci, 2000). Academic performance made contingent on external mechanisms, like close evaluation, inducements, or threats, reliably fails to tap into the inner resources of students (Reeve, 2006). Autonomous behavior like self-regulated learning flourishes in classroom environments that nurture generative psychological states of students (Niemiec & Ryan, 2009; Ryan & Deci, 2002).
Applications of self-determination theory to teaching have found that autonomy and competence-supportive teachers facilitate student internalization of school (Chirkov & Ryan, 2001; Reeve, 2002), build student self-esteem and efficacy (Deci, Vallerand, Pelletier, & Ryan, 1991), facilitate student engagement (Hardre & Reeve, 2003; Reeve, 2006; Reeve, Nix, & Hamm, 2003), increase interest in schoolwork (Tsai, Kunter, Ludtke, Trautwein, & Ryan, 2008), promote autonomous motivation (Reeve, Ryan, Deci, & Jang, 2008; Standage, Duba, & Ntoumanis, 2006), and increase achievement (Reeve, 2002). There is also evidence that controlling instructional practices undermine student interest in learning (Tsai et al., 2008), affect creativity and expression (Koestner, Ryan, Bernieri, & Holt, 1984), enhance anxiety, lead to more impoverished learning (Ryan & Weinstein, 2009), and diminish volitional action (Reeve, 2002).
We reason that the general propositions of self-determination theory supported by studies of teachers and classrooms are also true at the school level. Positive climates are just as likely to vary across schools as they do across classrooms. The difference between a classroom and school analysis is the nature of the phenomenon. School climate reflects shared perceptions of students toward the faculty, as well as shared perceptions of faculty toward students as a group. This is not the case in classrooms where the referent is the teacher and her instructional style. For schools, a supportive environment does not reflect individual behaviors or practices, but rather behavioral norms and regularities. Normative conditions are products of the historical relationships and social exchanges that unfold in schools. That is, norms define the shared beliefs and expected behavior of school members.
A self-regulatory school climate embodies teacherstudent interactions that produce trust and a deep commitment to academic excellence. Collective faculty trust in students is a consequence of engaged and responsible student behavior, and it underpins student-centered instructional practices. When faculty trust in students is normative, teachers are more attuned to student needs, listen to their concerns, express interest in their holistic development, and support their learning (Forsyth et al., 2011). Collective student trust in teachers reflects a historical pattern of teacher behavior that expresses concern for student development and well-being. Student trust establishes the relational infrastructure to transmit autonomy and competence-building experiences (Reeve, 2006). Academic emphasis reflects academic structures and processes that students perceive as furthering academic success and goal attainment. Together, facultystudent trust and academic emphasis form a climate that nourishes internal student regulation and self-directed action. Thus, we hypothesize that self-regulatory climate explains school-level differences in self-regulated learning.
The performance value of a healthy school climate is seldom challenged. Decades of school climate research have established school features like a connected community, cooperative interactions, safety, and supportive relationships as preconditions for learning and achievement (Cohen et al., 2009). In reference to the observable factors of a self-regulatory climate, considerable evidence links each unique condition to student and school achievement (Forsyth et al., 2011). Achievement effects attributed to faculty trust, student trust, and academic emphasis have been found even when controlling for contextual school conditions like poverty and prior achievement.
While evidence establishes a relationship between school climate and achievement, we argue the actual pathway is likely to be indirect, operating through student self-regulation. Student learning and achievement are partially dependent on autonomous motivation and self-regulated behavior (Niemiec & Ryan, 2009). Grolnick, Ryan, and Deci (1991) found that self-regulated elementary students had better achievement and adjustment in school than students with lower self-regulation. Vallerand, Fortier, and Guay (1997) found that autonomous motivation was predictive of the persistence and achievement of high school students. Autonomous motivation was also related to the academic performance, study strategies, and effort of medical students (Kusurkar, Cate, Vos, Westers, & Croiset, 2012). Internal motivation leading to self-regulated learning characterizes the type of inner resources energized by a self-regulatory climate. Collective faculty trust in students, collective student trust in teachers, and academic emphasis form a self-regulatory climate that underpins self-regulated learning. Student self-regulation for academic tasks is a psychological and behavioral state that influences student effort and academic performance. Thus, we hypothesize that self-regulated learning mediates the relationship between self-regulatory climate and math achievement.
Survey research was used to test the hypotheses. Data were collected in the spring of 2011 from teachers and students in 80 elementary and secondary schools from a large Southwestern urban school district. Student surveys were collected during the school day by the school district through designated school liaisons. Student participation was voluntary and parent assent was provided. Students were the unit of randomization. Thirty students were randomly sampled from either the fifth, seventh, ninth, or 11th grade depending on whether the school was an elementary, middle, or high school. Our selection of fifth, seventh, and 11th grades was guided by the need to randomly sample students who could accurately report on general teacher behaviors and school structures. Not accounting for mobility, fifth-grade students could have been in their elementary school for five years. Seventh-grade students were in their second year of middle school, and 11th-grade students were in the third year of high school. Ninth-grade students had been in their respective schools for less than a year. These students were sampled at the request of the school district.
Sampled students were randomly assigned to one of two surveys. The observable factors of a self-regulatory climate were included in survey form A, and those of self-regulated learning in form B. Surveys were separated to guard against common measurement bias in responses to survey items. Usable responses were received from 2,616 students, a 98 percent return rate. Researchers administered electronic surveys through Qualtrics to teachers in the 80 schools. Teachers were stratified by school, and then randomly assigned to one of two surveys. Usable responses were received from 1,039 teachers across the district, resulting in an overall teacher response rate of 68 percent.
Student achievement and demographic data came from the school district and state department of education. Math achievement comprised student scores from the state math exams for the 2,616 students who returned useable surveys. Achievement scores are scaled scores ranging from 4001000. The state sets 700 as the proficiency mark. Descriptive statistics in Table 1 describe characteristics of students and schools in the sample. Approximately 81 percent of the students qualified for the federal lunch program, and 47 percent were classified as minority students. Students missed an average of 8.87 days of school during the school year. They had an average math achievement score of 718. Schools in the sample had an average Free and Reduced Lunch (FRL) rate of 86 percent and an average minority rate of 68 percent.
Table 1. Descriptive Statistics for Student- and School-Level Variables
Note. N = 80 schools; N = 2,616; Free/reduced lunch reports the percentage of students in the sample who qualified for the federal lunch subsidy. Minority status reports the percentage of students in the sample classified as minority. Prior achievement was the average math achievement score for each school from the 2010 school year. The score is the average performance of all regular education students who completed the math curricular exam. Days absent was not related to the criterion variables so it was not included in the analytical models. Minority status was not related to self-regulated learning so it was not included in the model to test the first hypothesis. All scale measures were standardized to a mean of 0 and standard deviation of 1 in the multilevel analyses.
Collective Faculty Trust in Students
A subset of the Omnibus Trust Scale (Hoy & Tschannen-Moran, 1999) was used to measure faculty trust in students. The scale parallels the theoretical properties of collective trust in that it operationalizes teacher-shared perceptions of the openness, honesty, benevolence, reliability, and competence of students. Five items with a 6-point Likert response set ranging from strongly disagree to strongly agree make up the scale. These items were used because they had the highest factor loadings and represent the trust facets (Appendix A). Sample items include: students in this school can be counted on to do their work, teachers believe students in this school are competent learners, and teachers in this school trust their students.
Collective Student Trust in Teachers
Collective student trust was measured with the Student Trust in Teachers Scale (Adams & Forsyth, 2009). Like other trust measures, the student trust scale operationalizes trustworthiness through student-shared perceptions of the openness, benevolence, competence, honesty, and reliability of teachers. Five of the 13 items from the scale were used. These items were selected because they had the strongest factor loadings and captured the trust facets (Appendix A). The scale uses a 4-point Likert response set ranging from strongly disagree coded as 1 to strongly agree coded as 4. Sample items include: teachers are always ready to help at this school, teachers at this school really listen to students, and teachers at this school are good at teaching.
Student perceptions of academic emphasis were measured with five of the eight items from the Academic Emphasis Scale (Goddard, Sweetland, & Hoy, 2000). These items were selected because they had the highest factor loadings (Appendix A). The scale measures students views of their teachers efforts to push for higher levels of academic performance. Students report on teachers expectations of student effort and participation. High levels indicate that most teachers press all students toward academic achievement. The scale had a 4-point Likert response set ranging from strongly disagree coded as 1 to strongly agree coded as 4. Sample items include: this school has high expectations for student achievement, teachers in this school encourage students to keep trying even when the work is challenging, and teachers in this school place an emphasis on understanding school work not just memorizing it.
The self-efficacy for self-regulated learning scale was used to measure the metacognitive dimension of self-regulation (Bandura, 2006; Zimmerman, 1990; Zimmerman & Schunk, 2008). Seven items with a 4-point response set ranging from strongly disagree to strongly agree were used. Sample items include: I remember well information presented in class and textbooks, I arrange a place to study without distractions, and I get myself to study when there are other interesting things to do. Validity and reliability tests with data from this study showed good structural validity with factor loadings ranging from .67 to .74 and good reliability with an alpha of .89.
At the student level, FRL status was measured as a dichotomous variable with students qualifying for the federal lunch program coded as 1 and students not qualifying coded as 0. Minority status was also dichotomous with students classified as minority coded as 1 and non-minority students coded as 0. At the school level, FRL rate was measured as the percentage of students in a school qualifying for the federal lunch program. Minority rate was the percentage of students in a school classified as minority. The 2010 average math achievement score for each school was used as the measure of prior math achievement. Achievement scores are scaled scores ranging from 4001000. Prior achievement is the school average math score for all regular education students in the school who completed a math curricular exam.
The hierarchical data structure required us to follow an analytical process that first confirmed the theoretical conceptualization of self-regulatory climate, and secondly allowed us to test the hypotheses guiding the empirical investigation. To do this, we estimated within- and between-school variance in student and teacher perceptions of trust and academic emphasis. We also ran a confirmatory factor analysis on self-regulatory climate to examine the association of the observed factors. With support for the conceptualization of self-regulatory climate we were able to test our hypotheses through a model building process in HLM 7.0.
IntraClass Correlation Coefficients (1) and (2) were estimated to provide empirical support for the school-level nature of a self-regulatory climate. ICC results (Table 2) support the decision to specify self-regulatory climate as a school property. Variance in faculty trust, student trust, and student-perceived academic emphasis attributed to school differences was high and significant (FTS ICC(1) =.45; STT ICC(1) = .25; AE ICC(1) = .11). Approximately 9 percent of the variance in self-regulated learning and 19 percent of the variance in math achievement existed between schools. Within-group agreement among teachers and students was strong, as evidenced by the fact the ICC(2) estimates exceeded the .60 standard used to justify data aggregation (Cohen, Doveh, & Eick, 2001). Academic emphasis had relatively less variance attributed to school differences, but the variance still met accepted thresholds.
IntraClass Correlation Coefficients: Estimates of Group Dependence and Within-Group Agreement
Note. ** p < .01; Results confirm the school-level nature of the properties that make up self-regulatory climate. Results also report significant school-level variability in self-regulated learning but a weaker within-group agreement.
Next, we ran a confirmatory factor analysis on self-regulatory climate using AMOS 19.0. Factor analysis results support the theoretical alignment among collective faculty trust in students, collective student trust in teachers, and student-perceived academic emphasis. Factor loadings were statistically significant and ranged from .66 for collective faculty trust, .96 for collective student trust, and .83 for student-perceived academic emphasis (Figure 3). We created a composite variable by converting school scores for each factor into a standard score with a mean of zero and standard deviation of one. We then averaged standard scores to obtain a composite school-level value of self-regulatory climate.
Figure 2. Confirmatory factor analysis results for self-regulatory climate. Chi Square = 11.67 (p = .15), RMSEA = .042, CFI = .97, and TLI = .98.
Hypotheses were tested using HLM 7.0. For the first hypothesis, we fitted a Random Effects ANCOVA model where we controlled for FRL status of students. Preliminary analysis indicated that minority status and days absent were not related to self-regulated learning, so these variables were trimmed from the model. FRL status was grand-mean centered and fixed to the school level. Grand-mean centering has a computational advantage over no centering or group centering in that it reduces potential multicolinearity problems between intercepts and slopes across group estimations, and it isolates the net effect of school-level variables on an outcome by partialing out Level I effects (Raudenbush & Bryk, 2002). Contextual controls of percent of students qualifying for the FRL and percent minority were grand-mean centered at the school level.
SRLij = β0j + β1j (FRL) + rij
β0j = γ00 + γ01 (% FRL) + γ02 (% minority) + γ03 (SRC) + u0j
β1j = γ10
We used a 2-1-1 multilevel mediation process to test our second hypothesis. A 2-1-1 mediation is used when the antecedent variable is measured at the school level and the mediator and outcome variables are measured at the individual level (Krull & MacKinnon, 2001). Multilevel mediation analysis follows a similar process advanced by Baron and Kenny (1986) in that the first model tests the direct effect of the school-level predictor on the outcome variable. We tested a Random Effects ANCOVA in step one where we controlled for FRL and minority status at the student level. These Level I controls were grand-mean centered and fixed to the school level. At Level II we regressed average math achievement on FRL rate, percent minority, prior academic achievement, and self-regulatory climate. All Level II variables were grand-mean centered.
Mediation Step One
Math achievementij = β0j + β1j (FRL) + β2j (minority status) + rij
β0j = γ00 + γ01 (% minority) + γ02 (school FRL rate) + γ03 (prior achievement) + u0j
β1j = γ10
The second step of the mediation analysis was to test the relationship between the antecedent, self-regulatory climate in this case, and the mediator. For the second step, we used results from the model tested in our first hypothesis to evaluate the effects of self-regulatory climate on self-regulated learning. The analysis was based on the following equation.
Mediation Step Two
SRLij = β0j + β1j (FRL) + rij
β0j = γ00 + γ01 (% FRL) + γ02 (% minority) + γ03 (SRC) + u0j
β1j = γ10
The third mediation step tested the relationship between the antecedent and outcome variable with the mediator included (Krull & MacKinnon, 2001; Zhang, Zyphur, & Preacher, 2009). Again, we fitted a Random Effects ANVCOVA with FRL and minority grand-mean centered. Self-regulated learning was also entered as a Level I variable and grand mean centered to evaluate the change in the effect of self-regulatory climate on math achievement with the inclusion of the mediator. Equations for step three are presented below.
Mediation Step Three
Math achievementij = β0j + β1j (FRL) + β2j (SRL) + rij
β0j = γ00 + γ01 (FTC) + γ02 (school FRL rate) + γ03 (prior achievement) + u0j
β1j = γ10
β2j = γ10
Model fit was assessed for all models with the deviance estimate. Deviance accounts for the lack of fit between sample data and the theoretical model, and is best used when comparing the fit between two or more models (Luke, 2004). For our purposes, we were interested in fit improvement (i.e., reduction in deviance) between the main effect model and the mediation tests with self-regulated learning entered as a Level I predictor. We tested the variance-covariance structure of two models with the hypothesis-testing function in HLM 7.0. This test compares the difference in model deviance with the chi-square distribution (Luke, 2004). A significant chi-square suggests that one model fits the data better than the other.
We hypothesized that self-regulatory climate is related to school-level differences in self-regulated learning. As previously mentioned, school differences explained approximately 9 percent of the variance in self-regulated learning. Table 3 contains results of Model 1 and the full model. The difference between the models is that self-regulatory climate was not included as a predictor in Model 1 but was included in the full model. Model 1 explained approximately 39 percent of the between-school variance in self-regulated learning. Student FRL status had a significant, negative effect on self-regulated learning (β02 = -.18, p < .05), explaining approximately 3 percent of within-school variance. At the school level, prior math achievement (γ01 = .20, p < .01) had a significant and small effect on school differences in self-regulated learning but not FRL rate or percent minority. The reduction in deviance indicates that model fit was better for Model 1 compared to the unconditional model.
The inclusion of self-regulatory climate in the full model had two significant effects on the explained school-level variance in self-regulated learning. First, it increased the amount of explained variance from 39 percent to 96 percent. Second, it eliminated the unique effect attributed to prior achievement. Self-regulatory climate had the largest effect on self-regulated learning (γ04 = .28, p < .01) and accounted for nearly all of the variance attributed to school differences. Model fit, as measured by reduced deviance, was also best for the full model. The strength of the relationship between self-regulatory climate and self-regulated learning can be seen in Figure 4. Student perceptions of self-regulated learning were higher for both FRL and non-FRL students as self-regulatory climate in schools increased.
Table 3. HLM Results for Self-Regulated Learning
Note. N = 80 schools; ** p < .01; *p < .05. School-level predictors were standardized to a mean of 0 and a standard deviation of 1. Δ Deviance for model 1 presents the difference from the null model to model 1. Δ Deviance for the full model presents the difference from model 1.
Figure 3. Effect of self-regulatory climate on the perceived self-regulated learning of Free and Reduced Lunch (FRL) and non-FRL students.
Note: Values on the X and Y axes are ZScores with a mean of 0 and standard deviation of 1.
Our second hypothesis tested the pathway between self-regulatory climate and achievement. We hypothesized that self-regulated learning would operate as a mediating variable, explaining the reason for any achievement effect attributed to self-regulatory climate. Step one of the analysis was to test the relationship between self-regulatory climate and math achievement after controlling for student and school characteristics. Results of Model 1 report significant and negative effects of FRL status (β01 = -.36, p < .01) and minority status (β02 = -.29, p < .01) on math achievement, as well as a significant relationship between prior school achievement and math achievement (γ03 = .28, p < .01). Model 1 explained approximately 81 percent of the school-level variance in math achievement. The inclusion of self-regulatory climate changed the unique effects of the contextual controls and increased the amount of explained variance in student math achievement. Self-regulatory climate (γ04 = .20, p < .01) had the strongest unique effect on math achievement and reduced the variance attributed to prior achievement from 8 percent to 1 percent. The change in deviance indicates that the full model was the best-fitting model, explaining about 88 percent of the between-school variance in math achievement. Figure 5 illustrates the differential achievement effect of self-regulatory climate for FRL and non-FRL students.
Figure 4. Effect of self-regulatory climate on the math achievement of Free and Reduced Lunch (FRL) students and non-FRL students.
Note: Values on X and Y axes are ZScores with a mean of 0 and standard deviation of 1.
The second criterion for mediation, the relationship between self-regulatory climate and the mediator (self-regulated learning), was satisfied in the test of our first hypothesis. Self-regulatory climate had a strong, unique effect on self-regulated learning. In step three of the mediation analysis we entered self-regulated learning into the mediation model as a student-level variable. Self-regulated learning (β03 = .34, p < .01) had a stronger effect on math achievement than FRL status (β01 = -.29, p < .01) and minority status (β02 = -.29, p < .01). Further, its inclusion in the full model reduced the effect of self-regulatory climate from two tenths of a standard deviation to one tenth of a standard deviation, supporting the third criterion for mediation. It is interesting to note that inclusion of self-regulated learning decreased the size of the regression coefficients for FRL status and self-regulatory climate but not minority status, FRL rate, and prior achievement. In sum, results of all three mediation steps support the hypothesis that self-regulated learning mediates the relationship between self-regulatory climate and achievement.
Table 4. HLM Results for Math Achievement
Note. N = 80 schools; **p < .01; *p < .05. School-level predictors were standardized to a mean of 0 and a standard deviation of 1. Δ Deviance for model 1 presents the difference from the null model to model 1. Δ Deviance for the full model presents the difference from model 1. Δ Deviance for the mediation presents the difference from the full model.
A self-regulatory climate establishes predictable and cooperative interactions through shared influence and risk taking, reducing the dependence on external controls that constrain behavior and undermine autonomous action (Adams & Forsyth, 2013). Theoretically, such a social organization is compatible with student psychological needs. The empirical results of our study demonstrated that a self-regulatory climate has consequences for student self-regulated learning and math achievement.
SELF-REGULATORY CLIMATE AND SELF-REGULATED LEARNING
Results overwhelmingly support the proposition that school climates can nurture or suppress self-regulated learning. Even when controlling for FRL rate, percent minority, and prior achievement, self-regulatory climate explained about 90 percent of the school-level variability in self-regulated learning. Consistent with explanations found in self-determination theory, the combination of collective faculty trust, collective student trust, and student-perceived academic emphasis form a normative context that supports the internal regulation of both FRL and non-FRL students. When combining our findings with studies on autonomy and competence-supportive teaching, it is not unreasonable to claim that normative conditions capable of energizing psychological needs are social resources that urban schools can leverage for improved student motivation and performance.
Two additional findings from the analysis of the first hypothesis merit discussion. First, FRL rate and percent minority were not related to self-regulated learning. Ordinarily, non-significant findings do not receive much attention, but in this case it is worth noting that school-level compositional variables commonly associated with underperformance had no effects on school differences in student volitional beliefs and behavior. Given our sample of urban schools, we cannot rule out the possibility that significant relationships may exist in a more representative sample of urban, suburban, and rural schools. That stated, it is still the case that poverty levels and minority composition varied across urban schools in our sample, and these differences were not associated with differences in student regulation. We do not deny that high concentrations of poverty present challenges for schools, but school norms that form a self-regulatory climate are not predetermined by poverty or minority composition. Schools can foster internal regulation in students with a climate of trust and academic emphasis no matter the social context.
Second, we were surprised by the attenuation of prior school achievement when self-regulatory climate was entered in the full model. The effect of prior school achievement diminished by nearly three times its effect size in the first model. A reduction in explanatory power does not diminish the contribution of prior achievement; instead, it raises questions about its effect on structures, processes, and practices that build healthy school climates. We conjecture that better-achieving schools have established climates where cooperative interactions, shared responsibility, and academic emphasis are the norm rather than the exception. Past achievement that meets annual yearly progress may work to build trust and drive academic emphasis. Conversely, weak academic performance makes schools susceptible to external controls that can constrain teacher and student behavior, thereby depleting schools of social resources that build and sustain self-regulated learning.
SELF-REGULATORY CLIMATE AND ACHIEVEMENT
Consistent with self-determination theory, self-regulatory climate operated through the inner motivational resources of students to influence math achievement. All three criteria for mediation were satisfied. First, self-regulatory climate had a strong direct effect on math achievement. Even with conservative student and contextual controls in place, self-regulatory climate had the largest unique effect on school differences in math achievement, contributing about two tenths of a standard deviation increase in math achievement for every one standard deviation increase in self-regulatory climate. This difference amounts to 32 scale points for the average student in a school that is one standard deviation above the mean of self-regulatory climate and the average student in a school one standard deviation below the mean. Self-regulatory climate had performance benefits for both FRL and non-FRL students. In fact, FRL students in schools with a high self-regulatory climate outperformed non-FRL students in schools with a weak self-regulatory climate.
Second, the relationship between self-regulatory climate and self-regulated learning was established in the first hypothesis. Finally, the inclusion of self-regulated learning reduced the effect size of self-regulatory climate. Self-regulated learning was the strongest predictor of math achievement. Its achievement effect was larger than FRL and minority status at the student level and prior achievement, percent minority, FRL rate, and self-regulatory climate at the school level. With the criteria for medication confirmed, we conclude that the relationship between self-regulatory climate and math achievement was mediated by student self-regulated learning.
Urban schools with a self-regulatory climate are more capable of using relational support, persuasion, and influence to engage students in academic tasks. Schools with depleted trust and academic emphasis, on the other hand, depend on strong external controls to force compliance with rules and expectations. External control may produce desired behavior in the short term, but it often does so at the expense of energizing the internal capacity of students (Niemiec & Ryan, 2009). This is not the case with a self-regulatory climate. Collective faculty trust, collective student trust, and academic emphasis form a school environment where achievement is partly driven by students internal control of their learning.
IMPLICATIONS AND CONCLUSION
Motivation has long been viewed as an essential quality of students who experience academic and personal success (Reeve, 2002, 2006). In general, autonomously motivated students persist in school, are more interested in academic tasks, engage in learning activities, and perform better (Kenny, Walsh-Blair, Blustein, Bempechat, & Seltzer, 2010). Educators capable of tapping into student inner resources like motivation are often more effective in the classroom (Niemiec & Ryan, 2009). Applications of self-determination theory to teaching practices overwhelmingly conclude that autonomy-supportive, competence-supportive, and relational-supportive teaching styles nurture motivation and regulation (Reeve, Jang, Carrell, Jeon, & Barch, 2004). Our results suggest that schools have differential effects on the motivational resources of students as well, with self-regulatory climate being an essential social condition for autonomous regulation and achievement.
IMPLICATIONS FOR SCHOOL IMPROVEMENT
School climate can have differential effects on student self-regulated learning and achievement. School climates defined by high collective trust and student-perceived academic emphasis set the stage for interactions capable of promoting higher achievement by energizing motivational processes of students. Low trust and poor academic emphasis signal a social environment with limited capacity to meet the learning needs and opportunities of students. What do these results mean for a reform agenda that seemingly invests more resources on tangible interventions than intangible social conditions? We argue that investments in technical capacity need to be matched with comparable investments in building a self-regulatory school climate.
Schools with high collective trust and strong academic emphasis can use these conditions to their advantage as they implement new curricula, assessments, instructional technology, and other improvement strategies. Implementation processes most likely to result in changed behavior enable school professionals to learn from their experiences and to adapt practices in a manner that closes performance gaps and satisfies unique needs of individual students (Spillane, Reiser, & Reimer, 2001. Collective trust facilitates cooperative interactions under which information exchange, knowledge creation, and learning flourish. Low trust, on the other hand, diminishes the capacity of schools to learn and grow from the implementation of improvement tools (Adams & Forsyth, 2013). Collective trust coupled with student perceptions of academic emphasis signal a learning environment that is positioned to embrace new improvement tools as providing additional resources and opportunities to better prepare students for their future.
Schools deficient in conditions that form a self-regulatory climate face a more difficult improvement journey. For these schools, implementation of improvement tools or strategies can be an avenue to build a self-regulatory climate. Attention to collective trust and academic emphasis directs improvement efforts to the sources of motivation and high-quality performance. No meaningful and sustainable change in student performance can occur without a social environment that energizes student psychological needs (Niemiec & Ryan, 2009). The responsibility for ensuring that school structures, processes, and practices capitalize on the inner resources of students falls to school professionals. It is hard to foresee reduced achievement gaps if urban schools cannot produce the social and psychological resources under which motivation and performance prosper.
IMPLICATIONS FOR RESEARCH
This study had limitations that future research can address. First, being a quantitative analysis our findings fail to provide a rich description of schools with a strong or poor self-regulatory climate. With social resources being hard to foster, it will be important for future research to describe how structures, processes, and practices interact to build a self-regulatory climate. Particularly insightful would be descriptive evidence capturing how urban schools facing immense external pressure to raise achievement can buffer teaching and learning from controls that standardize and constrain behaviors.
Second, we see value in measuring the effect of self-regulatory climate on student autonomy, competence, and relatedness, as well as other positive psychological states like academic grit, optimism, and hope. Studies on the effect of self-regulatory climate on psychological states can test self-determination theory in school settings. As it stands, our study was not a full test of self-determination theory. We primarily used self-determination theory as a lens to better understand how self-regulatory climate shapes achievement. Future work can seek to unravel the dialectical process whereby social factors in schools interact with psychological needs to shape student behavior and achievement outcomes. Results of this study are a positive first step for future applications of self-determination theory to school effects research.
Third, our measure of self-regulatory climate combines shared experiences of teachers and students. Although there is value in combining teacher and student experiences into a composite indicator, faculty trust in students does not directly measure student perceptions of autonomy support. Our claim that a relationship exists between faculty trust and autonomy-supportive practices is based on theoretical, not empirical, evidence. Future research could also use longitudinal data to better understand the causal relationship between self-regulatory climate and autonomous motivation. With cross-sectional data we cannot rule out causal reciprocity between a school climate supportive of psychological needs and self-regulated learning.
Urban schools provided an ideal setting for the study because they generally face greater internal and external pressures to improve student performance. We reasoned that evidence of a self-regulatory climate effect in an urban context would suggest, as Honig and Hatch (2004) argue, that schools can respond to improvement pressures in ways that enrich student engagement and learning. One such positive response of schools appears to be the targeting of social conditions that support psychological needs of students. Schools organized in ways that build collective trust and emphasize academic excellence can regulate student learning in ways that leverage the natural capacity of students to flourish. The absence of trust and academic emphasis, on the other hand, can plunge schools into a cycle of controlling structures and practices that, as explained by self-determination theory, thwart student capacity to thrive academically.
Schools in our study experienced similar policies, student compositions, structures, and demands, yet differed in their levels of self-regulatory climate. As a result, schools with higher self-regulatory climate had students with greater self-regulated learning and higher average math achievement. We believe self-regulatory climate has value for educators seeking to provide equitable learning opportunities for all students and for researchers seeking to account for achievement differences attributed to schools. In both cases, self-regulatory climate advances a construct and measure that conceptualizes and operationalizes school-level support for psychological needs.
Adams, C. M., Ware, J. K., Miskell, R. C., Dollarhide, E. A., & Forsyth, P. B. (in press). Self-regulatory climate: A positive attribute of public schools. Journal of Educational Research.
Adams, C. M., & Forsyth, P. B. (2013). Revisiting the trust effect in urban elementary schools. The Elementary School Journal, 114(1), 121.
Adams, C. M., & Forsyth, P. B. (2009). The nature and function of trust in schools. Journal of School Leadership, 19(2), 126153.
Assor, A., Kaplan, H., & Roth, G. (2002). Choice is good, but relevance is excellent: Autonomy-enhancing and suppressing teacher behaviours predicting students engagement in school work. British Journal of Educational Psychology, 72, 261278.
Baron, B. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 11731182.
Bandura, A. (2006). Guide for constructing self-efficacy scales. In F. Pajares & T. C. Udan
(Eds.), Self-efficacy beliefs of adolescents (pp. 307338). Charlotte, NC: Information Age.
Chirkov, V. I., & Ryan, M. R. (2001). Parent and teacher autonomy-support in Russian and US
adolescents: Common effects on well-being and academic motivation. Journal of Cross-Cultural Psychology, 32(4), 618635.
Cohen, A., Doveh, E., & Eick, U. (2001). Statistical properties of the rwg(i) index of
agreement. Psychological Methods, 6(3), 297310.
Cohen, J., Mccabe, E., Michelli, N., & Pickeral, T. (2009). School climate: Research, policy,
practice, and teacher education. Teachers College Record, 11(1), 180213.
Connell, J. P., & Wellborn, J. G. (1991). Competence, autonomy, and relatedness: A
motivational analysis of self-system processes. In M. R. Gunnar & L. A. Sroufe (Eds.), Self processes and development: The Minnesota symposia on child development (Vol. 23, pp. 4378). Hillsdale, NJ: Erlbaum.
Darley, J. M. (2004). Commitment, trust, and worker expenditure in organizations. In R. M. Kramer and K. S. Cook (Eds.), Trust and district in organizations: Dilemmas and approaches (pp. 127152). New York, NY: Russell Sage Foundation.
Deci, E. L., Eghrari, H., Patrick, B. C., & Leone, D. R. (1994). Facilitating internalization: The
self-determination theory perspective. Journal of Personality, 62, 119142.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self determination in human
behavior. New York, NY: Plenum Press.
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), 227268.
Deci, E. L., & Ryan, R. M. (2008). Self-determination theory: A macrotheory of human
motivation, development, and health. Canadian Psychology, 49(3), 182185.
Deci, E. L., Vallerand, R. J., Pelletier, L. G., & Ryan, M. R. (1991). Motivation and education: The self-determination perspective. Educational Psychologist, 26(3), 325346.
Forsyth, P. B., Adams, C. M., & Hoy, W. K. (2011). Collective trust: Why schools cant improve without it. New York, NY: Teachers College Press.
Goddard, R. D., Salloum, S. J., & Berebitsky, D. (2009). Trust as a mediator of the relationships between poverty, racial composition, and academic achievement: Evidence from Michigans public elementary schools. Educational Administration Quarterly, 45(2), 292311.
Goddard, R. D., Sweetland, S. R., & Hoy, W. K. (2000). Academic emphasis of urban elementary schools and student achievement in reading and mathematics: A multilevel analysis. Educational Administration Quarterly, 36(5), 683702.
Grolnick, W. S., Ryan, R. M., & Deci, E. L. (1991). The inner resources for school achievement: Motivational mediators of childrens perceptions of their parents. Journal of Educational Psychology, 83, 508517.
Hardre, P. L., & Reeve, J. M. (2003). A motivational model of rural students intentions to persist in, versus drop out of, high school. Journal of Educational Psychology, 95(2), 347356.
Honig, M. I., & Hatch, T. C. (2004). Crafting coherence: How schools strategically manage multiple, external demands. Educational Researcher, 33(8), 16-30.
Hoy, W. K. (2012). School characteristics that make a difference for the achievement of all students: A 40-year odyssey. Journal of Educational Administration, 50(1), 7697.
Hoy, W. K., & Sabo, D. J. (1998). Quality middle schools: Open and healthy. Thousand Oaks, CA: Corwin Press.
Hoy, W. K., Tarter, C. J., & Woolfolk Hoy, A. (2006). Academic optimism of schools. In W. K. Hoy & C. Miskel (Eds.), Contemporary issues in educational policy and school outcomes (pp. 135156). Greenwich, CT: Information Age.
Hoy, W. K., & Tschannen-Moran, M. (1999). Five faces of trust: An empirical confirmation in urban elementary schools. Journal of School Leadership, 9, 184208.
Jang, H., Reeve, J., & Deci, E. L. (2010). Engaging students in learning activities: It is not autonomy support or structure but autonomy support and structure. Journal of Educational Psychology, 102, 588600.
Kenny, M. E., Walsh-Blair, L. Y., Blustein, D. L., Bempechat, J., & Seltzer, J. (2010).
Achievement motivation among urban adolescents: Work hope, autonomy support, and achievement-related beliefs. Journal of Vocational Behavior, 77(2), 205212.
Koestner, R., Ryan, R. M., Bernieri, F., & Holt, K. (1984). Setting limits on childrens behavior:
The differential effects of controlling versus informational styles on intrinsic motivation and creativity. Journal of Personality, 52(2), 233248.
Krull, J. L., & MacKinnon, D. P. (2001). Multilevel modeling of individual and group level
mediated effects. Multivariate Behavioral Research, 36(2), 249277.
Kusurkar, R. A., Cate, T. J., Vos, C. M., Westers, P., & Croiset, G. (2012). How motivation affects academic performance: A structural equation modeling analysis. Advances in Health Sciences Education, 18(1), 5769.
Leana, C. R., & Pil, F. K. (2006). Social capital and organizational performance: Evidence from urban public schools. Organization Science, 17(3), 353366.
Luke, A. D. (2004). Multilevel modeling. Thousand Oaks, CA: Sage.
Niemiec, C., & Ryan, R. M. (2009). Autonomy, competence, and relatedness in the classroom:
Applying self-determination theory to educational practice. Theory and Research in Education, 7(2), 133144.
Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data
analysis methods (2nd ed.). Thousand Oaks, CA: Sage.
Reeve, J. (2002). Self-determination theory applied to educational settings. In E. Deci and R.
Ryan (Eds.), Handbook of Self-Determination Research (pp. 183204). Rochester, NY: University of Rochester Press.
Reeve, J. (2006). Teachers as facilitators: What autonomy-supportive teachers do and why their
students benefit. The Elementary School Journal, 106(3), 225238.
Reeve, J., & Jang, H. (2006). What teachers say and do to support students autonomy during a
learning activity. Journal of Educational Psychology, 98(1), 209218.
Reeve, J., Nix, G., & Hamm, D. (2003). Testing models of the experience of self-determination in intrinsic motivation and the conundrum of choice. Journal of Educational Psychology, 95, 375392.
Reeve, J., Jang, H., Carrell, D., Jeon, S., & Barch, J. (2004). Enhancing students engagement by increasing teachers autonomy support. Motivation and Emotion, 28, 147169.
Reeve, J., Ryan, R., & Deci, E. L., & Jang, H. (2008). Understanding and promoting autonomous
self-regulation: A self-determination theory perspective. In D. H. Schunk & B. J. Zimmerman (Eds.), Motivation and self-regulated learning: Theory, research, and applications (pp. 223244). New York, NY: Erlbaum.
Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic
motivation, social development, and well being. American Psychologist, 55(1), 6878.
Ryan, R. M., & Deci, E. L. (2002). Overview of self-determination theory: An organismic
dialectical perspective. In E. Deci and R. Ryan (Eds.), Handbook of self-determination theory research (pp. 336). Rochester, NY: University of Rochester Press.
Ryan, R. M., Stiller, J. D., & Lynch, J. H. (1994). Representations of relationships to teachers, parents, and friends as predictors of academic motivation and self-esteem. Journal of Early Adolescents, 14(2), 226249.
Ryan, R. M., & Weinstein, N. (2009). Undermining quality teaching and learning: A self-
determination theory perspective on high-stakes testing. Theory and Research in Education, 7(2), 224233.
Spillane, J. P., Reiser, B. J., & Reimer, T. (2001). Policy implementation and cognition:
Reframing and refocusing implementation research. Review of Educational Research, 72(3), 387431.
Standage, M., Duba, J. L., & Ntoumanis, N. (2006). Students motivational processes and their
relationship to teacher ratings in school physical education: A self-determination theory approach. Research Quarterly for Exercise and Sport, 77(1), 100110.
Tsai, Y., Kunter, M., Ludtke, O., Trautwein, U., & Ryan, R. M. (2008). What makes lessons
interesting? The role of situational and individual factors in three school subjects. Journal of Educational Psychology, 100(4), 460472.
Vallerand, R. J., Fortier, M. S., & Guay, Frederic (1997). Self-determination and persistence in a real-life setting: Toward a motivational model of high school dropout. Journal of Personality and Social Psychology, 72(5), 1161-1176.
Vansteenkiste, M., Simons, J., Lens, W., Soenens, B., & Matos, L. (2005). Examining the
motivational impact of intrinsic versus extrinsic goal framing and autonomy-supportive versus internally controlling communication style on early adolescents academic achievement. Child Development, 76(2), 483501.
Zhang, Z., Zyphur, M. J., & Preacher, K. J. (2009). Testing multilevel mediation using
hierarchical linear models. Organizational Research Methods, 12(4), 695719.
Zimmerman, B. J. (1990). Self-regulated learning and academic achievement: An overview.
Educational Psychologist, 25(1), 317.
Zimmerman, B. J., & Schunk, D. H. (2008). Motivation: An essential dimension of self-
regulated learning. In D. H. Schunk & B. J. Zimmerman (Eds.), Motivation and self-regulated learning: Theory, research, and applications (pp. 130). New York, NY: Erlbaum.
Exploratory Factor Analysis Results of the Faculty Trust in Student, Student Trust in Teachers, and Student-Perceived Academic Emphasis