A number of assessment instruments for self-regulation have been developed in the past few decades. The significance of assessment to the current research field on self-regulation, as claimed by Boekaerts et al. (2006), is two-fold: assessment allows researchers to conceptualize the results from empirical studies, and it allows the concept of self-regulation to be operationalized.
Various types of assessment tools have been adopted in the investigation of second language acquisition, including the use of standardized instruments, self-report questionnaires, interviews, think-aloud protocols, and diaries, etc. The most frequently adopted instrument for assessing L2 learning strategy use in the past few decades has been the Strategy Inventory for Language Learning (SILL) developed by Rebecca Oxford (1990). The items in SILL involve five-point rating scales ranging from ‘never or almost never true of me’ to ‘always or always true of me.’ Tseng et al.
(2006) pointed out that the scale descriptors in SILL are behavioral items, and therefore cannot be assumed “a linear relationship between the individual item scores and the total item scores,” which results in the problem that the scale score is not psychometrically justified.
The most common measurement tool for self-regulated learning includes questionnaires and structured interviews (Zimmerman & Martinez-Pons, 1988). The construct of self-regulated learning has often been described as multi-dimensional, incorporating environmental, cognitive, affective, and behavioral elements. In recent years, the role of motivational and cognitive variables in self-regulation has drawn
30
much attention. Boekaerts et al. (2006) criticized that some of the previous studies seemed to have presented overlapping constructs and unclear concept definition in the assessment tools. Nonetheless, among the assessment tools developed by the previous researchers, one particular scale has been soundly constructed and has been widely adopted in self-regulation and motivation related research, that is, Motivated Strategies for Learning Questionnaire (MSLQ) developed by Paul Pintrich and his colleagues (Pintrich et al, 1991).
2.3.1 Motivated Strategies for Learning Questionnaire (MSLQ)
Current self-report questionnaires in the research field of language learning tend to ask respondents to generalize their learning action across situations as an underlying trait (Winne & Perry, 2000). The most prestigious instrument of such in the field of educational psychology is the Motivated Strategies for Learning Questionnaire (MSLQ) developed by Paul Pintrich and his colleagues at the National Center at the University of Michigan (Pintrich et al., 1991, Pintrich & Garcia1995).
According to Pintrich and his colleagues, the MSLQ was designed to assess the motivational orientation and learning strategy use of college students and from a social-cognitive perspective.
The MSLQ has been under continuous modification since 1986, when the first version was released. The final version of MSLQ is composed of two sections:
motivation and learning strategies. There are 81 items in total, each of them scored on a 7-point Likert scale ranging from 1= ‘not at all true of me’ to 7 ‘very true of me.’
The scale scores are constructed by computing the mean of the items in each subscale.
The motivational scales, composed of 31 items, are designed based on three motivational constructs (Pintrich, 1988a, 1988b, 1989): expectancy, value, and affect.
The concept of expectancy refers to learners’ beliefs in their ability to accomplish a
31
task, and it is measured with perceptions of self-efficacy and control beliefs.
According to Garcia (2005), the definition of self-efficacy in the MSLQ was adopted in a broader way compared with other measures (e.g., the LASSI; Weinstein, Zimmerman, & Palmer, 1988) in that it combines both learners’ expectancy for success (specific to task performance) and learners’ judgment of their ability to accomplish a task and confidence in their skills to perform a task. Control beliefs, on the other hand, refer to learners’ beliefs that learning outcomes are contingent on their own effort, rather than being influenced by external factors such as the teacher, the course content, or luck. The components of value draw attention on the reasons why learners engage in a certain academic task. Three subscales are included in the MSLQ to measure value beliefs: intrinsic goal orientation (learners’ focus on learning and mastery), extrinsic goal orientation (learners’ focus on grades and praise from others), and task value beliefs (learners’ judgments of how interesting, useful, and important the academic task is). To sum up, the motivation section assesses learners’ goals and value beliefs for a learning task, their self-perceived ability in achievement, and their test anxiety. The scales are labeled intrinsic goal orientation, extrinsic goal orientation, task value, control of learning beliefs, self-efficacy for learning &
performance, and test anxiety.
The learning strategies section of the MSLQ is formed on the basis of three types of scales: cognitive, metacognitive, and resource management. Cognitive strategies refer to learners’ use of strategies for the processing of information from texts or lectures. Metacognitive control strategies are related to the use of strategies that assist learners to control and regulate their own cognition, while resource management concerns learners’ use of strategies for controlling resources such as study environment and the use of others during the learning process. The learning strategy section is consisted of 50 items, which are further divided into two
32
subcategories: (a) ‘Cognitive and Metacognitive Strategies,’ including the subscales of rehearsal, elaboration, organization, critical thinking, and metacognitive self-regulation; and (b) ‘Resource Management Strategies,’ incorporating time and study environment, effort regulation, peer learning, and help seeking.
Studies done with the MSLQ have shown that the motivational subscales and the learning strategies subscales represent an empirically validated assessment tool for assessing learners’ motivation and learning strategies use, and the subscales also present reliable predictive validity (Pintrich et al., 1993; Pintrich & Garcia, 1995).
Pintrich and his colleagues concluded that the subscales in the MSLQ can be adopted together or singly, and it can be modified to fit different research needs. Based on this fact, it is claimed that the MSLQ is a soundly-constructed means for assessing learners’ motivation and learning strategies in the classroom.
2.3.2 Self-Regulating Capacity in Vocabulary Learning Scale (SRCvoc)
With the increasing interest in self-regulation from an educational psychology perspective, various measurement tools have been developed as an attempt to investigate the role of self-regulation in different domains of learning. However, to date, very few scales tend to assess self-regulation in second language learning. In fact, the only validated measure of self-regulation in second language learning that has been found in the literature is a self-report instrument developed by Tseng, Dörnyei, and Schmitt (2006) named ‘Self-Regulating Capacity in Vocabulary Learning Scale (SRCvoc).’
The SRCvoc was constructed under two objectives: first, Tseng et al. proposed that the instrument should target the learners’ trait of self-regulatory capacity rather than specific learning behaviors. Therefore, the self-report items in SRCvoc, following the step of the MSLQ, are ‘general declarations or conditional relations’
33
rather than ‘descriptions of specific strategic behaviors.’ Second, the theoretical background of the SRCvoc should be formed on the basis of the taxonomies developed by Dörnyei (2001), based on Kuhl’s (1987) and Corno and Kanfer’s (1993) taxonomies of action control strategies described as follows (cf. Tseng et al. 2006, p.85-86):
1. Commitment control, which helps to preserve or increase the learners’
original goal commitment (e.g. keeping in mind favorable expectations or positive incentives and rewards; focusing on what would happen if the original intention failed).
2. Metacognitive control, which involves the monitoring and controlling of concentration, and the curtailing of any unnecessary procrastination (e.g.
identifying recurring distractions and developing defensive routines;
focusing on the first steps to take when getting down to an activity).
3. Satiation control, which helps to eliminate boredom and to add extra attraction or interest to the task (e.g. adding a twist to the task; using one’s fantasy to liven up the task).
4. Emotion control, which concerns the management of disruptive emotional states or moods, and the generation of emotions that will be conducive to implementing one’s intentions (e.g. self-encouragement; using relaxation and meditation techniques).
5. Environment control, which helps to eliminate negative environmental influences and to exploit positive environmental influences by making the environment an ally in the pursuit of a difficult goal (e.g. eliminating distractions; asking friends to help and not to allow one to do something).
The results of a series of elaborate statistical analyses revealed that the SRCvoc reached satisfactory reliability and validity. The researchers indicated that the model of SRCvoc showed that self-regulation can be subscaled into five facets: commitment control, metacognitive control, satiation control, emotion control, and environment control. Although this measurement scale focuses on a specific domain of second language learning, i.e. vocabulary acquisition, with its high reliability and validity, the
34
researchers further suggested that the scale can be viewed as a precursor in assessing self-regulation in second language learning, and modification can be made to investigate other learning domains.