Self-underestimating students in physics instruction: Development over a school year and its connection to internal learning processes

https://doi.org/10.1016/j.lindif.2015.08.021Get rights and content

Highlights

  • We found a tripartite development of ‘self-underestimators’ over a school year.

  • 41% ‘remained self-underestimating’, 26% ‘improved’, and 33% ‘decreased’.

  • Improving students' self-concepts aligned with their high cognitive characteristics.

  • Self-underestimators' improvement was linked to higher internal learning processes.

  • Hence, self-underestimators can be supported through their internal learning.

Abstract

Students' individual cognitive and motivational–affective characteristics play an important role for successful learning but are also influenced by learning processes. For ‘self-underestimating’ students, high content knowledge is not met by an according self-concept of ability. This study investigates the development of N = 360 of these students over their ninth grade. Furthermore, it explores the connection between their development and their internal learning processes during physics instruction. Internal learning processes included the perceived fulfillment of their basic motivational needs, intrinsic motivation, and cognitive learning activity. Via latent class analysis, three developmental patterns were identified: students (1) whose self-concepts ‘improved’ and aligned with their high cognitive characteristics, (2) whose alignment ‘remained self-underestimating’, and (3) whose low self-concept prevailed and cognitive advancement ‘decreased’. Findings suggest that positive development is connected with higher internal learning processes. These results indicate that a positive development of ‘self-underestimating’ students is possible and should be supported by fostering internal learning processes.

Introduction

Students differ regarding their cognitive and their motivational–affective characteristics (Organisation for Economic Co-operation and Development (OECD) (OECD), 2014). Single characteristics such as the self-concept of ability play an important role as prerequisites for student learning (Riding & Rayner, 2001). Oftentimes, however, it is the interplay of these characteristics that makes a difference. Regarding self-concept, for instance, it is highly relevant if a low self-concept is connected with high or with low cognitive characteristics since the accuracy of this self-judgment also influences individual perception and behavior (Dunning, Heath, & Suls, 2004) and, for students, their learning (Dunlosky & Rawson, 2012).

Based on student profiles uncovered by Seidel (2006) for ninth-grade physics students, this study focuses on a particular combination of student characteristics as prerequisites for learning: students with a high level of physics knowledge, yet a low self-concept of ability regarding physics. This group of students is at risk of underestimating their abilities. From an educational perspective, this specific ‘self-underestimating’ profile is particularly interesting as it combines a favorable predisposition for teaching, the students' high prior knowledge, with a challenging characteristic, the low motivational predisposition of students. The group's especially high content knowledge indicates that they have high capacities for achieving which is one of the most prominent indicators for learning outcomes (Hattie, 2009). However, as Pintrich, Marx, and Boyle (1993) highlight, learning is not only “cold and isolated cognition”. Instead, motivational aspects are crucial for cognitive advancement. Individuals who underestimate their abilities might hold themselves back from engaging in critical situations for development and hence, often fail to live up to their potential (Elliot & Church, 2003). Their teachers therefore must provide support to foster internal learning processes that nourish the students' self-concept – for its own sake as an important outcome of education (Shavelson, Hubner, & Stanton, 1976) and to ensure cognitive development.

The focus on development of motivational–affective abilities in classrooms has long been a focus of research (e.g., Shavelson et al., 1976). However, to our knowledge, research has not focused on the development of students' low self-concept in the light of high content pre-knowledge as well as the development of this group's high cognitive abilities. Furthermore, studies have suggested that students' internal learning processes such as their perceived fulfillment of basic motivational needs, intrinsic motivation, and cognitive learning activity are positively connected to student development (Dweck, Mangels, & Good, 2004). Yet, it is unclear which internal learning processes are connected with the development of students whose subject-related self-concept needs special furtherance at the same time as their high content knowledge must be utilized and nourished.

In this study, we are interested in how students who have been characterized as ‘self-underestimating’ in physics instruction at the beginning of a school year develop over the course of one school year. Does their self-concept come to align with their high knowledge or does their learning suffer from their weak self-concept? Therefore, we analyze student characteristics data by means of latent class analysis (Hagenaars & McCutcheon, 2002). We want to know which latent classes of students who were characterized as ‘self-underestimating’ at the beginning of a school year can be separated at the end of the school year. Furthermore, this study attempts to shed light on the question if a potentially diverging development is connected with the students' internal learning processes. Thus, this study juxtaposes the perceived fulfillment of the three basic motivational needs, the intrinsic motivation, and the level of cognitive learning activity for the ‘development groups’ of ‘self-underestimating’ students and analyzes differences by nested analysis of variance.

Educational research agrees that cognitive and motivational–affective student characteristics are crucial for learning (Snow, Corno, & Jackson, 1996). Students need individual support according to their set of characteristics (Shuell, 1996). Recently, the analysis of a combination of different cognitive and motivational–affective student characteristics, like general cognitive abilities, interest, or self-concept has become the focus of studies like Hornstra, van der Veen, Peetsma, and Volman (2013) or Wormington, Corpus, and Anderson (2012). This contribution refers to a prior study by Seidel (2006) who indicated five different student profiles for physics instruction. The student profiles found at the beginning of the school year show differences in their cognitive abilities and in motivational–affective aspects (an overview can be found in Table 1 (values for MP 1)). Here, the term cognitive abilities refers to the students' general cognitive ability measured by a scale on reasoning abilities (e.g. figure analogies) and the students' subject knowledge assessed by a curriculum-adapted knowledge test developed for the study. Furthermore, we jointly refer to the students' interest in physics and their physics-related self-concept as the students' motivational–affective characteristics.

This study focuses on the development of students of the ‘self-underestimating’ profile that exhibits high cognitive abilities and low motivational–affective characteristics. 30% of students were classified as ‘self-underestimating’ at the beginning of the school year. Additionally, as two meaningful ‘reference groups’ of the Seidel (2006) study, we chose to consider the students that were classified in the overall favorable student profile at the beginning of the school year, the ‘strong’ students, and the overall concerning student profile, the ‘struggling’ students, in appropriate analyses. Values for their development at the end of the academic year and their internal learning processes during the school year were also calculated and considered to set the development and internal learning of the ‘self-underestimating’ group into perspective. Students with a ‘strong’ profile had shown high values both for cognitive as well as motivational–affective characteristics. They make up 24% of students at the beginning of the academic year. The 23% of all students who were profiled as ‘struggling’ at the beginning of the year had shown low values for all four cognitive and motivational–affective characteristics. Due to their fragmented nature, the remaining two profiles found in the study are not suitable as meaningful reference groups and hence, will not be presented or included in this paper.

Knowledge is both, the prominent outcome of learning processes and, by influencing how new information is processed and comprehended, crucial prerequisite for learning (Pintrich et al., 1993). The ‘self-underestimating’ students, who are in the focus of this study, possess a high level of physics knowledge. From this point of view, they are inclined to easily process and learn during the school year. However, as Pintrich et al. (1993) argue, cognition alone cannot satisfactorily describe learning. Instead, motivational beliefs are also essential. Academic self-concept is conceptualized in multiple models (for a review, see Marsh, 1990a). In this study, we refer to subject-specific self-concept as “a person's perception of himself” (Shavelson et al., 1976). It is influenced by an individual's experiences and environment. For a student's self-concept regarding physics as a school subject, the learning experiences and environment play an important role. Moreover, for the development of students, self-concept is not only a moderator variable for achievement but also an outcome of educational processes itself (Shavelson et al., 1976). For ‘self-underestimating’ students, the focus group of this study, a low self-concept is the other defining features. By investigating this specific group's development, this study sheds light on the interaction of high knowledge and a low self-concept.

In the present study referring to German physics instruction, a large group of students (on average 29% per classroom, but in some classrooms up to 65% of students) were classified into the ‘self-underestimating’ student profile (Seidel, 2006). Relatively more girls than boys exhibit this combination of cognitive and motivational–affective characteristics (Jurik, Gröschner, & Seidel, 2013). Physics as a subject may play a role. In schools, physics is perceived to be among the most difficult subjects only suitable for strong students (Osborne, Simon, & Collins, 2003).

Yet, in general, self-underestimation in students is a problem in education. On the one hand, since furthering self-concept is a major concern in any educational setting (Marsh & Hau, 2003). While for many individuals the self-concept aligns well with their cognitive abilities, there are individuals who are more likely to underestimate what they know or are capable of (Ackerman & Wolman, 2007). The inaccurate calibration of self-concept, i.e. when self-judgment does not reflect actual performance, can be grounded in information deficit or neglect, but also in uncertainty in the interpretation of feedback (Dunning et al., 2004). These issues that keep a student's self-concept low need to be addressed by teaching.

On the other hand, ‘self-underestimating’ students are at risk not to pursue new tasks and hold back from performing due to their negative self-image (Chiu & Klassen, 2010). Different psychological processes using unrealistically low expectations, worst-case scenarios, or effort withdrawal are said to be connected with underestimation of one's ability and reinforce the low self-esteem they originate from (Elliot & Church, 2003). Hence, a low self-concept is not only regrettable in itself but may actually hinder cognitive advancement.

In the complex interplay of achievement and self-concept (cf. Marsh, 1990b), this study is interested in how the development of both can be described for a group of students with a specific profile of high knowledge and low self-concept.

The cognitive and motivational–affective development of students can be influenced by instruction (Kunter et al., 2013). During instruction, individual student support by teachers monitors, fosters, scaffolds students motivational and cognitive internal learning processes (Pritchard, 2009). We consider internal learning processes to be the parts of students' learning that take place during instruction. In the current study, we focus on the connection between ‘self-underestimating’ students' development and their internal learning processes. In more detail, in this study, internal learning processes include students' perceived fulfillment of their basic psychological needs, intrinsic learning motivation, and cognitive learning activity.

The theory of self-determination, in general, investigates the conditions supporting or hindering “the natural processes of self-motivation and healthy psychological development” (Ryan & Deci, 2000). Hence, the concept of one's self is at the core of this psychological theory. For adolescents, a sense of autonomy and relatedness is shown to promote their self-esteem (Allen, Hauser, Bell, & O'Connor, 1994). In an educational setting, the fulfillment of students' basic psychological needs (sense of competence, autonomy, and social relatedness) as an internal learning process has been connected to what teachers do to support students (Reeve & Jang, 2006) and how students perceive themselves (Diseth, Danielsen, & Samdal, 2012). Additionally, a fulfillment of the basic motivational needs is strongly connected to students' sense of self and personality development (Deci and Ryan, 2008, Krapp, 2005). Therefore, teaching which addresses these psychological needs is assumed to have an impact on students' motivational characteristics (Pritchard, 2009, Rakoczy et al., 2008). Furthermore, intrinsic learning motivation, another motivational internal learning process, is said to be connected with a students' self-esteem (Pajares and Valiante, 1999, Praetorius et al., 2010). Overall, the fulfillment of basic psychological needs and experience of intrinsic learning motivation during instruction can lead to positive affective learning outcomes in students (Korthagen, Attema-Noordewier, & Zwart, 2014).

Besides motivation, students' cognitive learning activity is crucial for learning and development (Helmke, 2012). Teachers who use cognitive activation support their students' cognitive learning activities. In this study we refer to students' self-reported cognitive learning activity as “the kind of activity that really promotes meaningful learning” (Mayer, 2004, p.17), i.e. students' reported information processing in terms of basic elaborations, meaning to be able to follow the instructions of the teacher and to connect information to pre-experiences. Research shows that there is a link between basic cognitive activation and students' achievement (Baumert et al., 2010). Thus, cognitive learning activities are connected with cognitive development (Bransford, Brown, & Cocking, 1999).

Previous studies have already found that ‘self-underestimating’ students as an entire group report lower internal learning processes than other students. Students with a ‘self-underestimating’ profile report significantly lower intrinsic learning motivation and a lower level of cognitive learning activity than the average student (Jurik, Gröschner, & Seidel, 2014). Moreover, they also feel less socially related and less supported in their competence (Seidel, 2006). It is, however, not clear, if differences in internal learning processes within this group are connected with diverging development over the school year.

This study investigates the development of ‘self-underestimating’ students' characteristics and its connection to their internal learning processes in physics instruction. A schematic overview over variables and research questions is given in Fig. 1. This study attempts to answer the following research questions:

  • 1.

    How do ‘self-underestimating’ students develop over the course of one academic year?

    As the ‘self-underestimating’ group of students is defined by two characteristics, high knowledge and low self-concept, that are heavily interdependent, we expect three possibilities for development of this group: First, a group that does not change remarkably with respect to their combination of those characteristics. They gain knowledge over the school year solely due to their high prior knowledge without adjusting their self-concept. Second, we expect a group that develops their self-concept over the school year to match their high cognitive abilities. Third, we presume to find a group that suffers cognitively and fails to advance in their knowledge due to their low self-concept. Regarding the distribution of ‘self-underestimating’ students among those three groups and the amount of motivational increase or cognitive decline of the respective groups, we cannot make any assumptions based on the theory.

  • 2.

    Which internal learning processes of students in physics instruction are connected with ‘self-underestimating’ students' development?

    As our measures of internal learning are connected with both, cognitive and motivational–affective development, we expect the higher ratings for all internal learning processes for the group that we expect to remain cognitively advanced and, simultaneously, develop their self-concept to match their high knowledge, since they develop cognitively and motivationally. Especially their levels of intrinsic motivation and support of autonomy and social relatedness are expected to be higher for this group as they play a crucial role for motivational development. Furthermore, the group that falls behind cognitively and retains their low self-concept is expected to report the lowest level for all internal learning processes. We expect especially their level of cognitive learning activity to be lower than the other groups' as they do not develop cognitively. The group that, after the school year, still has high knowledge but a low self-concept is expected to report lower internal learning processes than the first group, but higher than the second.

Section snippets

Sample and design

In the present study, student questionnaire and test data of a German video study in Physics instruction are analyzed (for details see Seidel, Prenzel, & Kobarg, 2005). Our focus group of ‘self-underestimating’ students included N = 360 students from a randomly selected sample of NC = 50 ninth grade physics classrooms of highest and intermediate secondary school level with a total of NT = 1235 students (49% girls, 51% boys). All students were tested and asked about their content pre-knowledge,

‘Self-underestimator development groups’

The latent class analysis for student characteristics data of ‘self-underestimating’ students measured at the end of the school year revealed a tripartite development. One group of ‘self-underestimating’ students' showed an improvement over the academic year by evolving especially their self-concept dramatically and staying advanced in their level of knowledge, one group remained ‘self-underestimating’ with high knowledge but a low self-concept, and the third group exhibited a declining

Discussion

This study investigated the development of ‘self-underestimating’ students. The first research question was concerned with the change in their student characteristics, especially their physics knowledge and their self-concept. Our hypotheses supposed one group with an increase in self-concept along with a remaining high level of content knowledge, a group of students that maintained both, high knowledge and low self-concept, as well as a group with a drop in relative knowledge along with a

Conclusions

The findings of this study showed that, over the course of one school year, positive development was possible for students who were at risk of underestimating their abilities. With almost three fourths of ‘self-underestimating’ students still having a low self-concept (and low interest), however, it was not the majority of those students who exhibited a favorable development of their characteristics. Over 30% forfeited their cognitive advance and did not gain content knowledge over the school

Acknowledgments

We thank Richard J. Shavelson for his thoughtful comments on a preliminary version of this manuscript. The preparation of this manuscript was supported by a research grant of the German Research Foundation (DFG, SE 1397/7-1).

References (52)

  • W. Ahmed et al.

    A structural model of self-concept, autonomous motivation and academic performance in cross-cultural perspective

    Electronic Journal of Research in Educational Psychology 10

    (2006)
  • J.P. Allen et al.

    Longitudinal assessment of autonomy and relatedness in adolescent-family interactions as predictors of adolescent ego development and self-esteem

    Child Development

    (1994)
  • J. Baumert et al.

    Teachers' mathematical knowledge, cognitive activation in the classroom, and student progress

    American Educational Research Journal

    (2010)
  • J. Baumert et al.
    (1998)
  • E.L. Deci et al.

    Self-determination theory: A macrotheory of human motivation, development, and health

    Canadian Psychology

    (2008)
  • Å. Diseth et al.

    A path analysis of basic need support, self-efficacy, achievement goals, life satisfaction and academic achievement level among secondary school students

    Educational Psychology: An International Journal of Experimental Educational Psychology

    (2012)
  • D. Dunning et al.

    Flawed self-assessment: Implications for health, education, and the workplace

    Psychological Science in the Public Interest

    (2004)
  • C.S. Dweck et al.

    Motivational effects on attention, cognition, and performance

  • A.J. Elliot et al.

    A motivational analysis of defensive pessimism and self-handicapping

    Journal of Personality

    (2003)
  • J. Hattie

    Visible learning: A synthesis of over 800 meta-analyses relating to achievement

    (2009)
  • K.A. Heller et al.

    Kognitiver Fähigkeitstest für 4. Bis 12. Klasse, Revision

    (2000)
  • A. Helmke

    Unterrichtsqualität und Lehrerprofessionalität. Diagnose, Evaluation und Verbesserung des Unterrichts (4. Edition, Schule weiterentwickeln – Unterricht verbessern. Orientierungsband)

    (2012)
  • L. Hoffmann et al.
    (1998)
  • L. Hoffmann et al.
    (1997)
  • Cited by (13)

    • Opening ‘windows’ for teachers to change classroom discourse

      2020, Learning, Culture and Social Interaction
      Citation Excerpt :

      Students often receive evaluative feedback (Hattie & Timperley, 2007), which, if they give an incorrect answer, which can decrease their self-concept of ability. Accordingly, students with low self-concept of ability may be even less involved in classroom discourse, resulting in an increasing gap in learning development between students with a higher self-concept of ability and those with a lower self-concept of ability (Huber, Häusler, Jurik, & Seidel, 2015). To avoid this, it is important for teachers to be well aware of their students' characteristics, especially their self-concept of ability, and to investigate how students' self-concept of ability changes within the context of TPD dealing with productive classroom discourse.

    • Is teacher judgment accuracy of students’ characteristics beneficial for verbal teacher-student interactions in classroom?

      2018, Teaching and Teacher Education
      Citation Excerpt :

      In addition, students with high self-concept of ability more frequently receive supportive teacher feedback (Blumenfeld, Pintrich, Meece, & Wessels, 1982; Ismail & Majeed, 2011). In this sense, a so called ‘Matthew’-effect often occurs, resulting in an increasing gap in further learning developments between students with higher or lower characteristics (Huber, Häusler, Jurik, & Seidel, 2015; Lipowsky et al., 2007; Pauli & Lipowsky, 2007). The ‘Matthew’-effect originally describes the phenomenon that “the rich get richer and the poor get poorer” (Merton, 1968) and has been applied to the educational context in the way that “strong students get stronger and struggling students struggle even more” (Stanovich, 1986).

    • Underachievement in physics: When intelligent girls fail

      2016, Learning and Individual Differences
      Citation Excerpt :

      Such unfavorable conditions correspond with the underachieving girls' low self-concept and interest in physics and may contribute to the development of underachievement. Enhancing students' cognitive activity during physics learning, by means of teacher questions that stimulate deep-reasoning or teacher feedback, for example (Jurik, Gröschner, & Seidel, 2014), supports high ability students with a low self-concept in physics (Huber, Häusler, Jurik, & Seidel, 2015). More generally, cognitively activating instruction and unbiased assessment that focus more on conceptual understanding than on memorizing and practicing formulae application, link to prior experiences, address a broad range of problem contexts (e.g., from health physics to aeronautics), and support discussion can be expected to increase underachieving girls' self-concept and interest in physics and to positively influence performance (see Häussler & Hoffmann, 2002; Hoffmann, 2002; Hulleman & Harackiewicz, 2009; Lorenzo, Crouch, & Mazur, 2006; Siegle, Rubenstein, & Mitchell, 2014; Tho, Chan, & Yeung, 2015; Zohar, 2006; Zohar & Sela, 2003).

    View all citing articles on Scopus
    View full text