The Evolution of the Field of Learning Environments Research

Abstract

This article reviews half a century of remarkable expansion and internationalisation in the field of learning environments research, including milestones such as the creation of the American Educational Research Association’s Special Interest Group on Learning Environments and Springer’s Learning Environments Research: An International Journal. Several widely used, extensively validated and economical questionnaires for assessing learning environments (e.g., What Is Happening In this Class? WIHIC) are discussed. A review of research identifies how learning environment researchers have generated robust knowledge about how to improve student outcomes through creating positive learning environments; demonstrated the value of including learning environment dimensions as process criteria of effectiveness when evaluating educational programmes and teaching methods; and provided teachers with straightforward approaches to use in action research aimed at improving their students’ learning environments. To facilitate and motivate future research and practical applications, the 56-item WIHIC is included in an appendix, and some newer and emerging lines of research are identified (e.g., cross-national studies; the physical environments of educational buildings and learning spaces; and advances in methods of statistical analysis for learning environment studies).

1. Introduction

For this special issue of Education Sciences devoted to the topic of the effects of learning environments on students’ outcomes, I was asked to provide a historical perspective on the birth of the field and its subsequent evolution. In this article, discussion is structured under four headings of historical beginnings and milestones (Section 2), assessment of learning environments (Section 3), types of research on learning environments (Section 4) and discussion (Section 5).

2. Historical Beginnings and Milestones

My first introduction to the field of learning environments was in the early 1970s when I was undertaking research for my Ph.D. involving an evaluation of Australia’s first national curriculum project, namely, the Australian Science Education Project [1]. This research was guided and inspired by Herbert Walberg’s evaluation of Harvard Project Physics in the 1960s in the USA [2,3]. These two evaluations shared a common design involving a comparison of new and traditional curricula in terms of changes between pretest and post-test on a broad range of valued student outcomes (e.g., achievement, attitudes, inquiry skills, understanding of the nature of science).

Walberg’s evaluation of Harvard Project Physics was pioneering in that, in addition to the wide range of student outcomes, criteria of curricular effectiveness included the classroom learning environment. Subsequently, I included student perceptions of classroom environment in my evaluation of the Australian Science Education Project. This approach to assessing classroom processes through high-inference questionnaires responded to by participants was methodologically innovative at the time when the dominant approach involved low-inference classroom observations conducted by external observers.

A fascinating pattern in the evaluations of both Harvard Project Physics and the Australian Science Education Project was that the new and traditional curricula could not be distinguished in terms of student outcomes, but they could be differentiated in terms of the classroom learning environment created. Apparently, innovations in curricula and teaching methods initially have an impact on improving classroom environment and that it is only subsequently that these more positive environments lead to improved student outcomes.

I undertook a five-month sabbatical with Walberg at the University of Illinois in Chicago during the blizzard of 1979. During that year, I was introduced to the annual meeting of the American Educational Research Association (AERA), the world’s largest educational research conference. Soon, I became a regular participant in annual AERA conferences, which seemed to provide an excellent opportunity to disseminate my own research on learning environments, meet other learning environment researchers from around the world, and generally provide a catalyst for the spread of learning environment ideas and research around the world.

Although the AERA has over 100 Special Interest Groups (SIGs) on research on a very wide variety of topics, with each SIG being allocated its own program space/slots at AERA annual meetings, no AERA SIG at that time had learning environments research as a central focus. Therefore, in 1984, a major milestone in the evolution of the learning environment field was when Walberg and I (together with another colleague, Chad Ellett) decided to make an application to AERA to create a new AERA SIG called Learning Environments. When this application was successful, the SIG Learning Environments in 1985 started its own program at each AERA annual meeting with its own invited speaker, best paper award and opportunity to publish presentations in an annual refereed SIG-sponsored monograph. Today, this highly successful and international SIG has led to each AERA annual meeting program having over 100 presentations listed in the index under the heading ‘learning environments’. Wubbels [4] describes how attending AERA conferences nearly every year for him was “an inspiring event in my professional and personal life” and that the SIG Learning Environments “has been of the utmost importance in developing the program of research of our research group... and the dissemination of its results worldwide” (p. 20).

The annual monographs published by the SIG Learning Environments provided the basis upon which I was able to approach Kluwer Academic Publishers with the idea of starting up the field’s first journal. So, the next landmark in the evolution of the field of learning environments occurred in 1998 when the first volume of Learning Environments Research: An International Journal was published by Kluwer (now Springer Nature) (www.springer.com/journal/10984, accessed on 15 December 2022). In 2002, this journal was in its 25th volume, is rated in Scimago Quadrant 1 in multiple fields including Education, and is included in the Web of Science. Although I have served as an Editor-in-Chief of this journal since its inception, the journal’s international reach has been enhanced by always having one or more Regional Editors for each of North America, Europe and Australia/Asia.

The next landmark occurred in 2008, a decade after the birth of the journal, when I established the book series entitled Advances in Learning Environments Research published by Sense Publishers (now Brill), which I co-edited initially with Jeffrey Dorman and now with David Zandvliet. The first volume in this series was Outcome-focused Education: Determinants and Effects [5], several volumes have focused on physical environments [6], and the recent Thirty Years of Learning Environments: Looking Back and Looking Forward [7] provides celebratory reflections on the accomplishments of the AERA SIG Learning Environments during its first 30 years.

3. Assessment of Learning Environments

There is a wide range of research methods used to assess and investigate learning environments, including direct observations, interviews, questionnaire surveys, and case studies [8]. Tobin and Fraser [9] advocate mixed-methods approaches which combine qualitative and quantitative methods, and Aldridge et al. [10] provide a useful example of this combination in a cross-national study of learning environments in two countries. However, undoubtedly, the most frequently used way of assessing learning environments in prior research has been to use ‘high inference’ questionnaires that tap students’ and teachers’ perceptions and therefore characterise settings through the eyes of the participants themselves.

One of these questionnaires is the What Is Happening In this Class? (WIHIC) [10], which is currently the most frequently used learning environment questionnaire worldwide. The nature of the WIHIC is clarified in

Table 1. Scale Description and Sample Item for each WIHIC Scale.

Scale	Scale Description Extent to Which	Sample Item
Student Cohesiveness	students know, help and are supportive of one another.	1 make friendships among students in this class.
Teacher Support	the teacher helps, befriends, trusts and shows interest in students.	The teacher talks with me.
Involvement	students have attentive interest, participate in discussion, perform additional work and enjoy the class.	I give my opinions during class discussions.
Investigation	there is emphasis on the skills and processes of inquiry and their use in problem solving and investigation.	I explain the meaning of statements, diagrams and graphs.
Task Orientation	it is important to complete activities planned and to stay on the subject matter.	I know the goals of this class.
Cooperation	students cooperate with one another on learning tasks.	I work with other students on projects in this class.
Equity	students are treated equally by the teacher.	I get the same opportunities to answer questions as other students.

Items are scored 1, 2, 3, 4 and 5, respectively, for the responses of Almost Never, Seldom, Sometimes, Often and Very Often.

which lists the names of its seven scales together with a scale description and sample item for each scale. Items are answered using a five-point frequency response scale ranging from Almost Never (scored 1) to Very Often (scored 5). All of the WIHIC’s 56 items are listed in Appendix A.

The WIHIC’s wide usage for many purposes around the world is illustrated by Fraser’s [11] tabulation of 21 studies in 12 languages and in 13 countries (including the USA, UAE, Australia, Greece, India, Singapore, Indonesia, China, Korea and South Africa). This table captures major applications and selected results from the extensive research with the WIHIC, including a cross-national study of classroom environments in Taiwan and Australia [10]. More recently, Hanke and Fraser [12] compared middle-school students’ mathematics classroom environments in the USA and Hong Kong, whereas Long et al. [13] compared American preservice teachers’ perceptions of their learning environments before and after pandemic-related course disruption.

Significant research has been undertaken to modify the WIHIC for suitability for use in the contexts and languages of China and Greece. In China, the WIHIC was modified to form the NWIHIC (New What Is Happening in This Class?) and validated among 2280 grade 5–9 students from nine schools by Cai and colleagues [14]. The NWIHIC has six scales from the original WIHIC (Student Cohesiveness, Teacher Support, Involvement, Task Orientation, Cooperation and Equity) as well as the two new scales of Differentiated Instruction and Ongoing Assessment. Principal Component Analysis (PCA), Confirmatory Factor Analysis (CFA) and reliability analysis confirmed the NWIHIC’s validity. In Greece, Charalampous and Kokkinos [15] rigorously developed and validated a new elementary-school version of the WIHIC in the Greek language (the G-EWIHIC).

The WIHIC’s scales were included in the Constructivist-Oriented Learning Environment Survey (COLES), which was designed to provide feedback as a basis for reflection in teacher action research [16]. The COLES also includes the scales of Differentiation, Young Adult Ethos, Formative Assessment (extent to which students feel that their assessment tasks make a positive contribution to their learning) and Assessment Criteria (extent to which assessment criteria are explicit so that the basis for judgements is clear and public. For a sample of 2043 grade 11 and 12 students from 147 classes in nine Australian schools in Western Australia, the COLES displayed sound factorial validity and internal consistency reliability. The COLES was used successfully with teachers in conjunction with classroom observations and teacher collaboration in a whole-school professional learning initiative in Western Australia [17].

Researchers have modified and expanded existing learning environment questionnaire to better accommodate specific educational settings, such as computer-assisted classrooms [18] and science laboratories [19]. The Science Laboratory Environment Inventory (SLEI) includes unique dimensions such as integration, which focuses on the extent to which theory and laboratory instruction are integrated and which has been shown to be strongly linked with student outcomes. The SLEI was initially validated with 5477 high-school and university students in 269 classes in six countries, namely, Australia, the UK, USA, Nigeria, Canada and Israel [19].

Subsequently, the SLEI has been cross-validated and used in research in numerous countries. Lee et al. [20] validated a Chinese version of the SLEI in Taiwan and reported relationships between student self-efficacy and the levels of classroom Open-endedness and Rule Clarity. The SLEI also has been successfully validated and used in Korea [21], Israel [22] and Spain [23].

An extensive program of learning environment research that originated in the Netherlands involves the Questionnaire on Teacher Interaction (QTI), which assesses eight aspects of teacher behaviour based on the dimensions of influence (dominance–cooperation) and proximity (opposition–cooperation): Leadership, Helping/friendly, Understanding, Student responsibility/freedom, Uncertain, Dissatisfied, Admonishing, and Strict behaviour. In addition to the use of the QTI in the Netherlands [4,24,25], it also has been used in research in Brunei [26] and Indonesia [27].

The QTI has been adapted and used extensively among Chinese-speaking students. Sivan and colleagues [28] developed a Chinese version of the QTI (the C-QTI) and psychometrically validated it with students in Hong Kong. Using ‘smallest space analysis’, Sivan and Cohen [29] with 612 secondary students in Hong Kong demonstrated that C-QTI fits its underlying circumplex model for interpersonal teacher behaviour. An exceptionally rigorous development and validation of another Chinese version of the QTI involved 2000 secondary-school students from mainland China [30].

4. Types of Learning Environment Research

Section 4 describes three important types of learning environment research, namely, associations between the learning environment and student outcomes (Section 4.1), evaluation of educational programmes (Section 4.2), and improving learning environments (Section 4.3). Additionally, the discussion in Section 5 briefly identifies some emerging lines of learning environment research.

4.1. Associations between Learning Environment and Student Outcomes

A very strong tradition in learning environment research has involved investigating relationships between a range of student outcomes and the nature of the learning environment. This research has revealed consistent, strong and positive associations between the quality of students’ learning environments and their cognitive and affective outcomes. This line of research began using data collected as part of the previously mentioned evaluations of Harvard Project Physics [2,3] and the Australian Science Education Project [1]. A comprehensive meta-analysis involving 17,805 students revealed more-favourable student outcomes in learning environments with more cohesiveness, satisfaction and goal direction and less disorganisation and friction [31]. In a study of 1121 undergraduate students in Ethiopian universities, Tadesse and colleagues [32] found that students’ perceptions of constructivist learning environments (especially personal relevance) were associated with learning outcomes. When Wan [33] investigated the joint influence of classroom and family environments on critical thinking among 2189 secondary students in Hong Kong, classroom environment was a stronger predictor than family environment.

In order to obtain very large and high-quality data for research on outcome–environment relationships, some researchers have conducted secondary analysis (i.e., analysis for new purposes of existing data that were collected previously for different purposes). For example, Khine, Fraser and Afari [34] analysed PISA (Programme for International Student Assessment) data from 14,167 students in the United Arab Emirates to explore associations between three learning environment characteristics (cooperation, teacher support, investigation) and three non-cognitive student outcomes (epistemological beliefs, self-efficacy, attitudes). Using structural equation modelling, statistically significant associations were found between each learning environment scale and each non-cognitive outcome.

Fraser and Kahle [35] conducted secondary analysis of data collected from 7000 students in 200 schools as part of a Statewide Systemic Initiatives in Ohio. The study was unusual in that it encompassed three different types of environments (classroom, home, peer) and both achievement and attitude outcomes. Each of the three environments accounted for statistically significant amounts of unique variance in student attitudes, but only the class environment accounted for statistically significant amounts of variance in student achievement.

4.2. Evaluation of Educational Programmes

As noted in Section 2, Walberg’s historically important evaluation of Harvard Project Physics included learning environment dimensions among its criteria of effectiveness, thereby pioneering the use of learning environments assessments in many subsequent evaluations. For example, in Texas, Long et al. [13] traced changes in the learning environment among 230 preservice teacher education students during the switch to remote learning associated with the COVID-19 pandemic. These students perceived a statistically significant decline in student cohesiveness, teacher support, involvement, task orientation, and equity during the change to online learning, with the largest decline of 0.56 standard deviations occurring for student cohesiveness.

Flipped classroom instruction has been evaluated in terms of its effect on the classroom environment in Turkey [36] and the USA [37]. This American study revealed that flipped instruction was associated with greater cooperation and innovation but less task orientation, while the Turkish study suggested that flipped instruction was associated with higher student satisfaction and general belongingness.

Aldridge et al. [38] evaluated a national curriculum reform involving outcomes-based education (OBE) in South Africa using a school-level learning environment questionnaire with 403 teachers. It was found that teachers who had been involved with OBE perceived their school environments as having significantly more familiarity with OBE and work pressure.

It can be argued that the litmus test of the success of any professional development program for teachers is the extent of changes in teaching behaviours and ultimately student outcomes. In an evaluation of a two-year mentoring program for beginning elementary teachers in Florida, Pickett and Fraser [39] assessed teachers’ teaching behaviour in terms of their 573 students’ perceptions of classroom environment. The effectiveness of the mentoring program was supported by improvements over time in the learning environment, students’ attitudes and students’ achievement.

4.3. Improving Learning Environments

It is important not only that researchers conduct the sorts of learning environment studies reviewed in Section 4.1 and Section 4.2 but also that teachers draw on learning environment concepts and methods to improve their students’ learning environments. Teachers have successfully used feedback information based on assessments of their students’ perceptions of their actual and preferred learning environments to guide practical improvements in their classrooms. Earlier, these improvement attempts involved the simple steps of: assessment; feedback; reflection and discussion; intervention; and reassessment [40]. Over time, these simple steps have evolved and been expanded to: assessment and feedback; reflection and focusing; planning; implementing and refining; and reassessment [41]. Some case studies of teachers’ successful application of these change strategies in improving learning environments are reported below.

In South Africa, teachers applied these techniques for improving their classroom learning environments based on feedback from actual and preferred versions of the WIHIC. In the KwaZulu-Natal Province, 31 inservice teachers in a distance-education programme administered a primary-school version of the WIHIC in the IsiZulu language to 1077 grade 4–7 students [42]. Different teachers used feedback from the WIHIC with different levels of success in improving their classroom environments.

In New Zealand, when Taylor [43] used co-constructive learning strategies to reshape lessons, grade 9 science students perceived greater personal relevance and shared control in their classroom environments. In one South Australian school, a teacher of Italian used the COLES in her attempt to include her grade 8 and 10 students in co-construction [44]. Improvements in COLES scores over time, together with qualitative insights from interviews and narratives, supported the value of students co-constructing the learning environments of their classes.

In Australia, the COLES has been used in numerous action research studies involving teachers’ attempts to improve their learning environments. The first study established the validity of the COLES with a sample of 2043 grade 11 and 12 students in 147 classes in nine schools as well as using qualitative methods in case studies of how eight teachers used the questionnaire feedback [14]. Bell and Aldridge’s book [41] reported small but significant pretest–post-test changes over three years for a sample of 6107 students in 562 classes whose teachers implemented the approach to improving classroom environments. When Henderson and Loh [15] used the COLES in one school to guide teachers’ professional learning among 25 teachers and their 500 students each year, teachers highly valued this type of feedback.

5. Conclusions

During the half of a century since I first encountered the field of learning environments during my PhD research, this field has undergone remarkable expansion and internationalisation. Learning environment researchers have: developed and validated economical questionnaires that have been used extensively around the world; generated robust knowledge about how to improve student outcomes through creating positive learning environments; demonstrated the value of including learning environments dimensions as process criteria of effectiveness when evaluating educational programmes and teaching methods; and provided teachers with straightforward approaches to use in action research aimed at improving their students’ learning environments.

To facilitate and motivate future research and practical applications, this article has provided in Section 4 information about the world’s most widely used learning environment questionnaire, namely, the What Is Happening In this Class? (WIHIC), as well as listing its 56 items in Appendix A. This article also has described major applications of learning environment assessments in investigating associations between student outcomes and the learning environment (Section 4.1), in evaluating educational programmes (Section 4.2), and in teachers’ practical attempts to improve their learning environments (Section 4.3).

In seeking new insights, some researchers have conducted learning environment studies that cross international boundaries. For example, Hanke and Fraser [12] involved a large sample of 1309 grade 7 and 8 mathematics students in the USA and Hong Kong in responding to four learning environment scales (teacher support, involvement, cooperation, equity) and some attitude scales. Students in the USA perceived their learning environments significantly more positively than did Hong Kong students, but Hong Kong students enjoyed their classes more than American students. Interestingly, Aldridge et al. [10] reported the same pattern of results in their study of junior high-school science students in Australia (n = 1081) and Taiwan (n = 1879): although Australian students perceived their classroom environments more positively than Taiwanese students, Taiwanese students enjoyed their science classes more than Australian students.

Recently, there has been a resurgence in interest in the physical environment of educational buildings and learning spaces [6] in contrast to the psychosocial characteristics of learning environments reviewed in the present article. For example, researchers have focused on flexible furniture in the USA [45], a ‘deskless’ school in Finland [46], LED lighting in a pre-K child development laboratory in the USA [47], and a comparison of round-shaped and crescent-shaped seating arrangements in Japan [48]. However, it is disappointing to observe that although new designs or redesigns of learning spaces usually aim to change the psychosocial learning environment in specific ways, it has been rare to evaluate their effectiveness in terms of changes in psychosocial characteristics.

Based on a comparative ethnographical approach, Kokko and Hirsto in Finland investigated processes by which physical spaces can be transformed into learning environments [49]. Baars and colleagues [50] proposed a model for understanding and investigating the supportiveness of physical environments to innovative pedagogies, stressing the need for a psychosocial–physical relational approach to designing new learning environments.

Over the past decade, learning environment research methods have undergone impressive changes to more explanatory, predictive and model-testing approaches according to den Brok et al. [51]. These advances include approaches to statistical analysis such as confirmatory factor analysis, multilevel analysis, structural equation modelling and Rasch scaling. Recently, Kuzle [52] proposed a novel approach to obtaining elementary-school children’s insights into psychosocial aspects of their classroom environments via qualitative content analysis of their drawings.

As this paper demonstrates, the learning environments field had experienced impressive expansion and diversification during the previous half century. Two major catalysts for this expansion have been the American Educational Research Association SIG (Special Interest Group) on Learning Environments and Learning Environments Research: An International Journal. As this special issue of Education Sciences on the topic of learning environments research illustrates, the field is ripe for further exciting new developments in the coming decades.