Boosting Students’ Academic Performance: Trait-State Model of Academic Achievements in Higher Education



Yüklə 207,59 Kb.
səhifə1/4
tarix23.11.2017
ölçüsü207,59 Kb.
#12177
  1   2   3   4

CHAINED MEDIATION MODEL OF ACADEMIC PERFORMANCE


Author's Accepted Manuscript (AAM)

Rogaten, J., & Moneta, G. B. (in press). Positive and negative processes underlying academic performance: A chained mediation model. Journal of Happiness Studies.

DOI: 10.1007/s10902-016-9765-6

Submitted: August 29, 2015

Resubmitted: March 31, 2015

Accepted: May 28, 2016

Published online: June 6, 2016


Positive and negative structures and processes

underlying academic performance: A chained mediation model
Jekaterina Rogaten

University of the Arts, London, UK

Giovanni B. Moneta

London Metropolitan University, London, UK


Running head: CHAINED MEDIATION MODEL OF ACADEMIC PERFORMANCE
Total number of words: 11,908
Submission: August 29, 2015

Revision: March 31, 2016

Authors’ Note

Correspondence should be addressed to Giovanni B. Moneta, London Metropolitan University, School of Psychology, Room T6-20, Tower Building, 166-220 Holloway Road, London N7 8DB, United Kingdom, telephone +44 (0)20 7133-2573, email g.moneta@londonmet.ac.uk.



Abstract

This study proposed and tested a comprehensive, chained mediation model of university students' academic performance. The hypothesized model included adaptive-positive and maladaptive-negative submodels. The structures and processes in the adaptive-positive submodel were hypothesized to facilitate students’ academic performance, whereas the structures and processes in the maladaptive-negative submodel were hypothesized to undermine it. A sample of 373 university students completed a set of questionnaires measuring their approaches to studying, positive and negative affect, evaluation anxiety, use of creative cognition, motivational orientations, and adaptive and maladaptive metacognitions. Participants’ end-of-semester and prior semester academic performance was retrieved from the university registry. A structural equation model explained 90% of the variance in students’ future academic performance, supported all but one hypothesized intermediate paths, and revealed that only positive affect in studying and prior academic performance predict directly future academic performance. The theoretical and practical implications of these findings are outlined.


Keywords: academic performance, approaches to studying, evaluation anxiety, metacognition, motivation, negative affect, positive affect, university students, use of creative cognition.

Introduction

Research within Higher Education has traditionally been concerned with psychological variables that undermine learning, and has paid less attention to psychological variables that facilitate learning. For example, there are over 200 studies that looked at the effect of evaluation/test anxiety on learning (Zeidner, 1998), and only a handful of studies that looked at the effect of positive emotions on learning (e.g., Pekrun, Frenzel, Goetz, & Perry, 2007; Pekrun, Goetz, Titz, & Perry, 2002). There has also been a scarcity of empirical research that considers simultaneously the effects of psychological variables that undermine and facilitate learning. In order to start filling this gap, the present study developed and tested a comprehensive, chained mediation model of academic performance that examines simultaneously positive-adaptive and negative-maladaptive structures and processes underlying learning. It is hoped that the development of such a modeling approach will help to prevent educational interventions from being biased toward either negative-maladaptive or positive-adaptive psychological variables, and to identify the primary target variables for interventions aimed at improving students’ chances of success in Higher Education.

The present study chose academic performance as the measure of learning for the following reasons: (a) it is universally recognized as the most appropriate measure of learning, (b) it is free from self-report biases, (c) it allows a direct comparison of research findings with the results of other studies, and (d) standardized assessments allow direct comparisons between students (e.g., Anaya, 1999; Bowman, 2010; Gonyea, 2005). Educational research has consistently found that prior academic performance is the strongest predictor of future academic performance (e.g., Diseth, 2007; Duff, 2004; Zeegers, 2004). Therefore, the present study examined the effects of psychological variables on future academic performance controlling for the effect of prior academic performance. The inclusion of past performance as a control variable in the model implies that the outcome variable of this study is progress (or regression) in academic performance relative to students' prior academic performance.

The empirical studies conducted to date have identified a large number of psychological variables that correlate with academic performance in Higher Education (e.g., see reviews by Hattie, 2009, and Richardson, Abraham, and Bond, 2012). The large majority of these variables are specific to the educational context (e.g., use of study aids, attitude toward studying, or test-taking strategies). Departing from such a context-specific approach, the present study selected only key context-specific psychological variables (i.e., approaches to studying and evaluation anxiety) and included instead a set of general psychological variables that have been researched systematically in a wide range of fields of psychological enquiry – in particular, personality and social psychology, work psychology, and clinical psychology – and that are closely linked to general theories of psychological functioning. This was done based on the belief that at this stage of scientific development there is a strong need of integrating educational psychology with other major fields of psychology. This integrative approach was pursued in prior studies (e.g., Rogaten, Moneta, & Spada, 2013; Spada & Moneta, 2012, 2014) by using subsets of the variables considered here. Most of the relationships investigated in the present study have already been tested in a piecewise fashion on different student samples. The present study seeks the development and testing of a comprehensive model that integrates all the hypothesized relationships.

From the personality psychology perspective, all psychological variables can be viewed as either processes or structures. Processes are variables that signify the variability of behavior within a person in response to different situations, whereas structures are variables that signify the similarity or average tendency of an individual’s behavior across situations and times (e.g., Fleeson, 2001; McAdams, 1995). All the variables selected for the model of this study were classified as either processes or structures based on their underlying theory and the properties of their measurement scales, chiefly stability and domain dependence. The core modeling principle adopted in this study is that processes should be the direct predictors of academic performance, whereas structures should influence academic performance indirectly, by influencing processes.

Typically, the variables and hypotheses constituting a chained mediation model are presented starting with the predictors (structure variables), followed by the first-order mediators (process variables), and ending with the second-order mediators (process variables). However, in the hypothesized model of this paper the structure variables have numerous relationships with other variables in the model. Therefore, when stating the hypotheses for the structure variables one would need to define all the process variables that are predicted by them. Therefore, for the sake of clarity and conciseness we will construct the model in a backward fashion, first considering the process variables and then the structure variables.



Processes that predict academic performance

Approaches to studying

Students’ Approaches to Learning (SAL) theory (Entwistle, Hanley, & Hounsell, 1979) states that there are three main approaches to studying: deep, strategic, and surface. These can be broadly categorized into adaptive and maladaptive. Deep and strategic approaches to studying are characterized by deep interpretation and analysis of new information and by target-oriented attitudes toward learning, respectively, and hence are adaptive. Surface approach to studying is characterized by rote-learning and shallow understanding of study material (Entwistle & Peterson, 2004), and hence is maladaptive. A study that administered measurement scales derived from SAL theory and other theoretical perspectives found strong convergence among the items from the different scales that transcend theoretical differences and consolidate the description of the three approaches to studying (Speth & Brown, 2011): (a) the deep approach implies an intention to understand, personalize, and integrate the information being learned with prior knowledge; (b) the strategic approach implies a tendency to integrate from the start the information being learned with contextual cues, particularly those concerning assessment; (c) the surface approach implies a tendency to reproduce unselectively the learning material without personal involvement, and to ignore contextual cues.

Students’ approaches to studying may change as a result of a change in the educational environment, particularly in assessment methods (e.g., Kember & Gow, 1994; Marton & Säljö, 1976b), or following study skills interventions (Norton & Crowley, 1995; Solomonides & Swannel, 1995), and hence qualify as process variables. Consistent with SAL theory, strategic approach and, to a lesser extent, deep approach positively correlate with academic performance, whereas surface approach negatively correlates with academic performance (e.g., Byrne, Flood, & Willis, 2002; Diseth, 2007; Diseth, Pallesen, Brunborg, & Larsen, 2010). Therefore, it was hypothesized that:

H1: (a) Strategic and (b) deep approaches to studying will positively correlate with academic performance, whereas (c) surface approach to studying will negatively correlate with academic performance.

Affect

Affect is the most general and primitive construct in emotional research (Russell, 2003) and is a conceptual umbrella for both moods and emotions (Wyer, Clore, & Isbell, 1999). Positive affect includes emotions like love, interest, contentment, whereas negative affect includes emotions like anger, fear, and disgust (Fredrickson, 1998).

Three theories of emotions provide suggestions on how to position positive and negative affect within the model of academic performance. First, control-process theory (Carver & Scheier, 1990, 2001) postulates that affect works as a signal of progress and a regulator of effort in achievement endeavors, in such a way that if progress is faster than desired, the individual will experience positive affect, whereas if progress is slower than desired, the individual will experience negative affect. Second, the mood-as-input theory (Martin, Ward, Achee, & Wyer, 1993) posits that affect guides the start and stop mechanisms of intentional behavior as follows: (a) positive mood prolongs engagement with an activity if the objective in pursuing the activity is enjoyment, and shortens engagement if the objective is goal attainment; (b) negative mood shortens engagement with an activity if the objective is enjoyment, and prolongs engagement if the objective is goal attainment. Finally, the broaden and build model of positive emotions (Fredrickson, 1998, 2001) postulates that positive emotions broaden thought-action repertoires through enhancing cognition (Fredrickson & Branigan, 2005; Fredrickson & Joiner, 2002) and expanding the scope of attention (Gasper & Clore, 2002) (broaden hypothesis), and that positive emotions – even short-lived ones – have long-term positive effects by increasing physical, psychological, and social resources (Cohn, Fredrickson, Brown, Mikels, & Conway, 2009) (build hypothesis).

The stability of the positive and negative affect students experience in studying across consecutive two semesters was found to be moderate (Rogaten & Moneta, 2015b), indicating that affect in studying is a process variable. Consistent with the outlined theories of emotions, students' positive affect in studying was found to correlate positively with academic performance, whereas negative affect was found to correlate negatively (e.g., Dosseville, Laborde, & Scelles, 2012; Rogaten, Moneta, & Spada, 2013). Therefore, it was hypothesized that:



H2: (a) Positive affect in studying will positively correlate with academic performance, whereas (b) negative affect in studying will negatively correlate with academic performance.
Creative thinking

Creativity was identified as an important process in learning and a strong correlate of positive affect (e.g., Baas, De Dreu, & Nijstad, 2008; Isen, Daubman, & Nowicki, 1987). Following recent developments in creativity research, we were interested in examining the effect of context-dependent use of creative thinking in studying. We previously argued that creative ability and context-dependent use of creative cognition are related but distinct constructs (Rogaten & Moneta, 2015a, 2015b). Although a certain level of creative ability is needed in order to deploy creative cognition, it is possible that some students high in creative ability do not typically use their creative cognition in study contexts, whereas some students low in creative ability do (Rogaten & Moneta, 2015b). The following cognitive processes related to creativity were identified: divergent and convergent thinking, metaphorical and analogical thinking, perspective taking, and imagery (for a review see Davis, 2004). Keeping in mind that everyone can use creative cognition more or less effectively in their studying, but not necessarily do so, the use of creative cognition in studying is a process variable.

Students who use their creative cognition in studying should understand the subject matter better and learn faster, and hence experience more positive affect in studying. In support of this conjecture, the use of creative cognition in a semester was found to predict positive affect in the following semester (Rogaten & Moneta, 2015b). The relationship between the use of creative cognition and approaches to studying has not yet been investigated. Nevertheless, the use of creative cognition should promote the development of adaptive approaches to studying. For instance, metaphorical and analogical thinking and perspective-taking, which facilitate manipulation and transformation of ideas that result in new knowledge (Davis, 2004), are likely prerequisites of deep processing of information, and hence should facilitate the adoption of the deep approach to studying. Moreover, divergent and convergent thinking are two thinking strategies that enable individuals to come up with multiple ideas, and then narrow down the selection to one idea that meets the requirements of the problem at hand. This type of target-oriented thinking is a distinct characteristic of strategic learners (Entwistle & Peterson, 2004; Ramsden, 1979). As such, the use of creative cognition in studying should also facilitate the adoption of the strategic approach to studying. Therefore, it was hypothesized that:

H3: Use of creative cognition in studying will positively correlate with (a) positive affect in studying, (b) strategic and (c) deep approaches to studying.

Evaluation anxiety

Evaluation anxiety has been traditionally regarded as the key affective process variable undermining learning (e.g., Fleeson, 2001; McAdams, 1995; Spielberger & Vagg, 1995). Evaluation anxiety is anxiety that is specific to the situations where one’s performance can be negatively evaluated by others (Geen, 1991), and is an umbrella term for different types of anxiety, such as test anxiety, statistical test anxiety, and performance anxiety (Skinner & Brewer, 1999; Zeidner & Matthews, 2005). Evaluation anxiety and its different sub-forms were found to undermine cognitive efficiency – particularly by reducing working memory and attention – and academic performance (see review by Zeidner, 1998). Evaluation anxiety was consistently found to correlate with surface approach to studying (Cermakova, Moneta, & Spada, 2010; Moneta, Spada, & Rost, 2007; Spada & Moneta, 2012, 2014). Finally, test-anxious students were found to experience higher levels of negative emotions, particularly shame and guilt (Arkin, Detchon, & Maruyama, 1982; Stowell, Tumminaro & Attarwala, 2008). Therefore, it was hypothesized that:



H4: Evaluation anxiety will positively correlate with (a) surface approach to studying and (b) negative affect in studying.

Structures that predict processes

Motivational orientations

Self-determination theory (Deci & Ryan, 1985; Ryan & Deci, 2000) postulates that motivation is the core process and structure underlying learning. Intrinsic motivation is the tendency to engage in tasks because one finds them interesting, challenging, and enjoyable, whereas extrinsic motivation is the tendency to engage in tasks because of task-unrelated factors such as anticipation of rewards, surveillance, and competition (Deci & Ryan, 1985; Ryan & Deci, 2000). Intrinsic and extrinsic motivation were originally conceptualized as state variables that change across situations and times and are incompatible with each other at any given time. Amabile and co-workers (Amabile, Hill, Hennessey, & Tighe, 1994) have later defined and operationalized intrinsic and extrinsic motivation as independent traits to be driven either by the engagement of work or by a means to some end that is external to the work itself.

From its inception, SAL theory posited that the deep approach to studying is driven to intrinsic motivation (derived from interest in the subject matter), whereas the surface approach to studying is driven by extrinsic motivation (derived from an inner pressure to memorize the unconnected details of the subject matter), and found evidence of these associations in factor and cluster analyses of items measuring states (Entwistle & Wilson, 1977). When measured as traits, intrinsic motivation was consistently found to correlate with the deep approach to studying, whereas extrinsic motivation was found to correlate with the surface approach to studying (e.g., Moneta & Spada, 2009; Prat-Sala & Redford, 2010; Spada & Moneta, 2012, 2014). Therefore, it was hypothesized that:

H5: Trait intrinsic motivation will positively correlate with (a) deep approach to studying, whereas trait extrinsic motivation will positively correlate with (b) surface approach to studying.1

Intrinsic motivation implies appreciation of complexity, including task novelty, as an opportunity to explore, play with ideas, and acquire mastery (Deci & Ryan, 1985), and hence should foster the use of creative cognition. Extrinsic motivation energizes behavior by arousing ego-involving anticipations of success or failure (Deci & Ryan, 1985), and hence should foster evaluation anxiety. Consistent with these arguments, trait intrinsic motivation was found to correlate with creative output in a wide range of tasks (e.g., Amabile, Hennessey, & Grossman, 1986; Hennessey et al., 1989), and with the use of creative cognition in studying (Rogaten & Moneta, 2015a), whereas trait extrinsic motivation was found to correlate with evaluation anxiety (Spada & Moneta, 2012, 2014). Therefore, it was hypothesized that:



H6: Trait intrinsic motivation will positively correlate with (a) use of creative cognition in studying, whereas trait extrinsic motivation will positively correlate with (b) evaluation anxiety.

General metacognitions

Metacognition is a multidimensional construct (e.g., Antonietti, Ignazi, & Perego, 2000; Wells, 2002) that encompasses psychological structures, beliefs, and control functions that enable an individual to interpret and modify one's own thinking (Flavell, 1979). Metacognition is essential for determining what strategies one can use to perform any learning task, such as identifying required skills, detecting potential obstacles, assessing time and effort costs, and estimating potential benefits (Antonietti et al., 2000). Metacognition in the educational context typically refers to a form of higher order thinking characterized by the ability to self-regulate cognitive processes in learning. Such processes include identifying effective ways of carrying out a task, monitoring comprehension, and assessing learning progress after completing a learning task (Schraw, 1998).

From a personality psychology perspective, metacognitions are relatively stable traits and can be broadly separated into adaptive and maladaptive, in that they either facilitate or hinder problem solving in challenging situations (Beer & Moneta, 2010, 2012). On the one hand, adaptive metacognition is theorized to foster flexible switching of attention from a perceived threat to the task at hand based on the strategic demands of the situation, agentic search for alternative pathways, and flexible goal restructuring (Beer & Moneta, 2010). When adaptive metacognition is activated, the thinking becomes flexible and adaptable. On the other hand, maladaptive metacognition is theorized to foster excessive threat monitoring, perseverative thinking, and maladaptive coping in response to external stimuli and to one’s own internal states, and to maintain psychological dysfunction through these processes (Wells & Matthews, 1994; Wells, 2000). When maladaptive metacognition is activated, the thinking becomes negative, cyclical, and rigid.

The two metacognitions are likely to influence affect in studying and approaches to studying in diametrical ways. On the one hand, adaptive metacognition comprises the meta-emotions of interest and curiosity (Mitmansgruber, Beck, Höfer, & Schüßler, 2009) in one’s own primary emotional responses to a challenging endeavor, in such a way that difficult tasks are construed as positive challenges (Beer & Moneta, 2010). Therefore, adaptive metacognition should foster positive emotions by enhancing the appraisal of difficult learning tasks. In support of this conjecture, adaptive metacognition was found to correlate with positive affect in studying (Moneta, 2012) and work (Mackay & Moneta, in press). The relationship between adaptive metacognition and approaches to studying has not yet been investigated. Nevertheless, the ability to set flexible and feasible study goals and to set the mind for problem solving should foster deep and strategic approaches to studying.

On the other hand, maladaptive metacognition implies a focus on negative emotions and presumed or real environmental threats that trigger those negative emotions, rather than a focus on the task at hand (Wells & Matthews, 1994). Therefore, maladaptive metacognition should foster coping with negative emotions, with the consequence that leaving the real-world problem unattended typically results in even more negative emotions. Consistent with theory, maladaptive metacognition was found to foster negative emotions (Moneta, 2011) and to exacerbate the effect of perceived stress on negative emotions (Spada, Nikčević, Moneta, & Wells, 2008). Moreover, the excessive preoccupation with internal states and external threats should deplete attentional resources, and hence lead students to adopt a surface approach to studying. Indeed, maladaptive metacognition was consistently found to correlate with surface approach to studying (Moneta et al., 2007; Spada & Moneta, 2012, 2014; Spada, Nikcevic, Moneta, & Ireson, 2006). Therefore, it was hypothesized that:

H7: Adaptive metacognition will positively correlate with (a) positive affect in studying, (b) strategic and (c) deep approaches to studying, whereas maladaptive metacognition will positively correlate with (d) surface approach to studying and (e) negative affect in studying.

The two metacognitions are also likely to have diametrical consequences on the use of attentional resources in studying. On the one hand, adaptive metacognition should foster a focus on the learning task as an opportunity to perform creatively, and hence foster the use of creative cognition. In support to this reasoning, adaptive metacognition was found to correlate with the use of creative cognition in studying (Rogaten & Moneta, 2015a). On the other hand, maladaptive metacognition should foster a focus on the threat from anticipated failure to perform satisfactorily, and hence foster evaluation anxiety. Indeed, maladaptive metacognition was consistently found to correlate with evaluation anxiety (Spada & Moneta, 2012, 2014; Spada et al., 2006). Therefore, it was hypothesized that:



H8: Adaptive metacognition will positively correlate with (a) use of creative cognition in studying, whereas maladaptive metacognition will positively correlate with (b) evaluation anxiety.

Goals of the study

Figure 1 shows the hypothesized chained mediation model of academic performance. The model comprises adaptive-positive and maladaptive-negative submodels. The adaptive-positive submodel includes psychological variables that are expected to foster academic performance, whereas the maladaptive-negative submodel includes psychological variables that are expected to undermine academic performance. The model contains 21 paths, of which 17 received some empirical support in prior studies and four are tested for the first time in this study. Previous studies tested only subsets of the variables and paths shown in the model. Testing the whole set of paths in a single model provides the opportunity to rule out spurious associations that may have received support because relevant competing variables were not controlled for. As such, the goals of the study are: (a) to test the model as a whole and each of its hypothesized links, and (b) to compare the two submodels in their ability to explain students' academic performance.

--------------------------------------

Insert Figure 1 about here

--------------------------------------

Method


Yüklə 207,59 Kb.

Dostları ilə paylaş:
  1   2   3   4




Verilənlər bazası müəlliflik hüququ ilə müdafiə olunur ©genderi.org 2024
rəhbərliyinə müraciət

    Ana səhifə