In social sciences, an event may be studied when a concept is operationalized; through that transformation, a theoretical hypothesis may be transformed into an empirical one (Gavard-Perret et al., 2008). Mackenzie notes that appropriate definitions should specify the constructs’ conceptual domain, be clear and definitive, consistent with existing scholarship, and be distinguishable from other constructs (MacKenzie, 2003). The definition (MacKenzie, 2003) and the measurement (Churchill, 1979) must first be determined to operationalize the construct.

Nielsen (2014) suggested that each theory has two components of equal importance: measurement and description (i.e., definition). He criticized management scholars for being “predominantly preoccupied with the latter – often at the expense of the former” and encouraged scholars to “pay more attention to measurements issues in general, and construct validity, appropriateness of measurement across contexts, as well as the matching of measurement to theory” (p.405). The operationalization of a concept, in this case, TE, allowed for its empirical testing (Gavard-Perret et al., 2008); thus, defining the construct is imperative when contributing to scholarship (MacKenzie, 2003). Churchill (1979) suggested that a valid construct should specify its domain, such that a definition involves “attributes of objects that are measured and not the objects themselves” (p.65). A good construct definition should specify its conceptual theme, provide clear, unambiguous terms, be consistent with previous scholarship, and differentiate itself from related constructs (MacKenzie, 2003).

Despite the importance and age of the concept, there has yet to be a consensus on how to define TE (Mastrokoukou et al., 2022). Young and Shaw (1999) observe the lack of clarity regarding a single, generally agreed-upon definition of TE. In Profiles of Effective College and University Teachers, the authors discuss “whether it [TE] might be defined in multiple ways, and if so, how it might be defined” (p.670). Yet, Young and Shaw’s methods focused solely on defining TE from the perspective of student evaluation of TE, thus capturing only one aspect of measuring TE.

When seeking to determine what TE is, O’Neill (1988, p.162) suggested researchers may look to the elements that contribute best to student learning, specifically the “teaching methods, under what conditions, with what students, in what subject areas, and at what grade levels contribute most effectively to student achievement.” There appears to be a common acceptance amongst scholars that a positive correlation exists between TE and student learning (Ding and Sherman, 2006).

Liu et al. (2022) did not define TE, though did reference their prior works about how to teach more effectively. Barnes (2000) used operational definitions to define TE: “Keeps an effective lesson pace. Minimizes time between events, makes smooth transitions,” “Supports and encourages the best efforts of students Maintains a positive tone in the class/lesson,” and “Controls and varies speech speed The speech speed is not static” [Hamann and Baker (1998), as cited by Barnes (2000, p.6)]. In addition, Barnes measured TE without adequately defining the concept. She is not alone in this approach, as researchers have published measures (of TE) without defining them (Becker, 2006; Blanchard et al., 2010).

There has been a lack of consensus in the literature on measuring TE. “Teacher quality is a complex phenomenon, and there is little consensus on what it is or how to measure it” [Lewis (1999), para. 3, as quoted by Stronge et al. (2011)]. Some research has been published with a combination of measures (“multiple measurements”) (e.g., Cinnamon et al., 2021; Lohman, 2021; Taylor et al., 2021), while some papers chose not to measure TE even though they were writing about TE (e.g., Voogt et al., 2013; Huang et al., 2015).

The lack of defining TE and the inconsistency in measuring may have harmed TE research’s progress and quality. It is through this paper we argue for pairing an existing strong definition of TE with educator-focused (input) and student-focused (outcome) measures of TE.

To collect our data, we applied a practical screen. Due to practical limitations (e.g., the first author’s language proficiency), only papers published in English were included. As the research aimed to conceptualize TE, we focused our search on scholarly peer-reviewed publication journals. Subsequently, we searched for literature using the Web of Science (WoS), Scopus, Education Resource Information Centre (ERIC) and Google Scholar, focusing on papers published from 1980 to 2023 to retrieve high-quality peer-reviewed journal articles containing “teach* effective*” or “effect* teach*” in the title, abstract, or keywords which were either the most cited or most cited in recent years. The asterisk represents the software’s ability to search any key term that includes “teach” when paired with “effective.” Figure 1 outlines our data collection process. We received 5,259 results from WoS. After filtering for journal articles, there were 3,056 journal articles and no non-English publications to remove. We sorted by most relevant and recently published and removed duplicates, collecting 109 peer-reviewed publications. To add robustness to our corpus, we added the most relevant results from Scopus, ERIC, and Google Scholar for 143 peer-reviewed papers.

Data extraction followed. We leveraged the extraction process (Okoli, 2015a) used. First, the primary author reviewed all papers collected, extracted the definitions of TE, and recorded how each study measured TE. To complete this extraction, the full publications were reviewed for the definition of TE; the abstracts, then the entire publication (when necessary), were reviewed for their measurement of TE. Where present, we extracted their TE definitions and measurements. The measurement types were then categorized into themes and re-coded using these themes.

The RA independently replicated this process with one exception: they received a list of coded themes to document the measurement types. The co-author “arbitrated” any differences between the primary author and RA in the papers’ extracted definitions and measurement categorizations. For example, some papers had partial TE definitions or defined a related construct (e.g., learning effectiveness). Only once two research team members (i.e., the primary author and RA, or one of the primary author or RA and co-author) agreed with the TE definition and measurement extraction (or exclusion) from each paper were the results finalized. Figure 2 outlines our data extraction process.

During our analysis, we evaluated TE definitions from our extracted data and looked at the relative representation of TE measures in papers that defined TE. Due to the variance in how TE was defined in the literature, we sought guidance from how the English language defines a definition. We generously interpreted the word “definition” from Merriam-Webster: “a statement of the meaning of a word or word group or sign or symbol”¹ and the Britannica Dictionary: “an explanation of the meaning of a word, phrase, etc.”²

To determine the quality of our findings of TE definitions in our target papers, we applied MacKenzie’s (2003) applicable three criteria of a “good definition” (specified conceptual theme, clear terms, and consistent with previous research), which were clearly distinguished from other constructs, as summarized in Table 1. We used the SCImago³ (SCI) Journal Rankings to understand the area of discipline of the journal and analyzed the definitions according to their foci (educator and/or student). We synthesized the studies and wrote our review in this paper’s Results and Discussion sections. Following argument-crafting guidance from Okoli (2015b), we wrote our discussion to offer a theoretical contribution to the literature.

In our corpus, we found only nine of the 143 papers defined TE, though some papers discussed constructs close to TE, such as teacher effectiveness (Goldhaber and Anthony, 2007 p.135; Sass et al., 2012 p.43), learning effectiveness (Kuo et al., 2022; Vlachogianni and Tselios, 2022; Walinski et al., 2023) and academic performance (Gumasing et al., 2023). Some studies discussed learning experiences (Campbell et al., 2004), and found TE was dependent, in part, on (an educator’s) personality traits (Renaud and Murray, 1996), while others analyzed student personality in relation to TE (Yunker, 1983). Thus, our review demonstrates that effective teaching is multidimensional, consistent with Pan and colleagues, who said, “understanding the multidimensionality of effective teaching is essential when validating instruments and interpreting final ratings of different teaching approaches” (Pan et al., 2021 p.3).

In our literature review, the concept of teacher effectiveness emerged as a related but different concept. Sass et al. (2012 p.431) define teacher effectiveness as “…teacher effectiveness can therefore be interpreted as the impact of a teacher on student learning gains…” while Goldhaber and Anthony (2007), define teacher effectiveness as “teachers’ contributions toward student gains in achievement.” Both definitions focus on the impact of one type of teaching intervention (i.e., the teacher) on student learning. Definitions of teacher effectiveness observed in our corpus are like the TE definitions observed. However, with the distinction of the teacher (in teacher effectiveness) as the antecedent to learning (versus a broader, less defined antecedent of learning when we observed TE definitions). While a teacher and their effectiveness may always appear to be integrated with TE directly, an organization or program may evaluate TE using both the TE construct and its teacher effectiveness sub-dimension.⁴ Thus, Teacher effectiveness may be a sub-dimension of TE. As such, we have analyzed both TE and teacher effectiveness in Table 1.

We noted that papers that defined TE and teacher effectiveness were from the Education (4), Economics and Econometrics (3), and Physical Therapy, Sports Therapy and Rehabilitation (2). Curiously, of the papers that defined TE and teacher effectiveness, only Galbraith et al. (2012), Rink (2013), and Adnot et al. (2017) quote other papers’ definitions of TE and teacher effectiveness, while Goldhaber and Anthony (2007), Sass et al. (2012), Philip and Garcia (2013), Reynolds et al. (2014), and Boring et al. (2016) all developed their own definitions of TE and teacher effectiveness. Table 1 provides a summary of definitions collected.

There exists a variety of TE definitions in the literature. Galbraith et al. (2012, p.357) quotes McKeachie (1979, p.385) when defining TE: “We take teaching effectiveness to be the degree to which one has facilitate student achievement of education goals.” Ward (2013, p.431) says, “teaching effectiveness is defined in terms of student learning.” This perspective leads to more questions because while many authors attempted to say what learning was (i.e., growth, achievement), Ward did not.

When evaluating the foci, we noted the TE definitions had both student and educator foci (Galbraith et al., 2012; Ward, 2013; Boring, 2017). Because Goldhaber and Anthony (2007), Sass et al.'s (2012), Philip and Garcia (2013), Rink (2013), and Adnot et al. (2017) definitions were for a similar construct (teacher effectiveness), we classified their foci as “not applicable,” while Reynolds et al. (2014) did not have a focus.

After applying Mackenzie’s (2003) criteria in Table 1, we observed one of the nine definitions, Galbraith et al.’s (2012) paper quoting McKeachie (1979), was rated higher, in some instances significantly higher than the other definitions in our corpus, suggesting that McKeachie proposed the most appropriate definition of the TE construct.

TE can be measured using student evaluation of teaching effectiveness (SETE), objective measures (objective), self-assessment, administrator evaluation (administrator), and peer review.

Student-focused measures for TE include SETE (Wilson, 1986; Renaud and Murray, 1996) and student objective evaluation. Objective measures include class means (McKeachie et al., 1971; Ni, 2013). Some papers measure TE with single measure of SETE (Hiebert et al., 2007; Allgood et al., 2015), while other papers include SETE as a part of multiple measures (Boring, 2017), use SETE to develop a conceptual model (Hargreaves, 1999) orin a case study (Esquembre, 2004). The papers analyzed that measured TE solely by SETE included various contexts such as business education (Paul et al., 2012), e-learning (Wang, 2003), and medical education (Rosen et al., 1993; Moraros et al., 2015).

Objective measures of TE include observer measures (Howes, 1997; Rink, 2013), proxies of care through subject improvement (Suri and Schultz, 1999; Ramani and Leinster, 2008), post-test results (Holcomb et al., 2002; Blanchard et al., 2010), and other objective measures of academic achievement (Hallinger et al., 2014; Bowers, 2016). Specific measures include the Protocol for Language Arts Teaching Observation (Grossman et al., 2014), student academic and behavioral outcomes (Espelage et al., 2013), changes between measurements using pre- and post-test results (Holcomb et al., 2002; Blanchard et al., 2010), and test scores and attendance (Holcomb et al., 2002).

The educator-centered measurements include administrator evaluation, self-assessment, and peer review. Administrator evaluation as a measurement of TE was used in Medical Interviewing and Interpersonal Skills Teaching in United-States Medical-Schools-Progress, Problems, and Promise (Novack et al., 1993), and in combination with other measures including objective (Jacob and Lefgren, 2008; Staiger and Rockoff, 2010), self-assessment (Price, 2012; Sobaih et al., 2016), and both peer review and objective (Fryer, 2014). Similarly, both self-assessment and peer-review measures are paired with other measures of TE. Self-assessment has been measured alongside SETE (Butler, 2012), objective (Rockoff et al., 2011; Hamre et al., 2012), and administrator evaluation (Price, 2012; Roeser et al., 2012), and peer review has been paired with administrator evaluation and objective measures (Fryer, 2014). No researchers in our corpus had used the TE measures self-assessment and peer-review on their own to measure TE, though some (Chen et al., 2021; Taylor et al., 2021; Hamann et al., 1998; Lohman, 2021; Baum and Brown, 1980; Centra, 1994) use these evaluations in combinations with SETE. Consequently, we consider that these measures, taken alone, measure Teacher Effectiveness which is another concept.

TE has been measured using a combination of student- and educator-focused measures, such as administrator evaluation and student objective measures (Jacob and Lefgren, 2008; Roeser et al., 2012). Some papers have chosen not to measure TE, including in review articles (Ashton, 1984; Reynolds et al., 2014), whilst developing a conceptual model (Srinivasan et al., 2011), and in a case study (Esquembre, 2004).

Table 1 outlines the variations of papers that define TE, whether each paper leveraged existing literature or created their own definition, an evaluation of the strength of their definitions, and how each paper that defined TE measured TE. For example, Galbraith et al.’s (2012) TE definition had both student and educator foci and measured TE using SETE, while Ward’s (2013) TE definition had the same dual foci and used student objective measurements for TE.

We observed three papers have both student and educator foci (Galbraith et al., 2012; Ward, 2013; Boring, 2017), one paper appears to not have a focus while five papers were deemed to not be applicable as they defined teacher effectiveness (Goldhaber and Anthony, 2007; Sass et al., 2012; Philip and Garcia, 2013; Rink, 2013; Adnot et al., 2017).

Also outlined in Table 1, SETE represented 22.2% of papers, Student Objective Measures were at 77.8% while 11.1% of papers that defined TE did not measure it. Of note, 11.1% of papers contained multiple measures of SETE and Student Objective Measures, hence our representation of measurements exceeds 100%.

Contexts in our observations included Higher Education (Galbraith et al., 2012; Boring et al., 2017), Unspecified (Philip and Garcia, 2013; Ward, 2013), and Primary and/or Secondary Education (Goldhaber and Anthony, 2007; Sass et al., 2012; Rink, 2013; Reynolds et al., 2014; Adnot et al., 2017).

Existing literature appears to be conflicted when determining how definitions and measurements of TE are paired to create a construct. As such, we conclude with an evaluation of the TE construct where we argue for pairing an existing strong definition of TE with educator-focused (input) and student-focused (outcome) measures of TE.

The prevailing omission when discussing and measuring TE without defining it could be due to “Obliteration by Incorporation” or OBI. Merton (1988, p.622) defines OBI as: “the obliteration of the source of ideas, methods, or findings by their being anonymously incorporated in current canonical knowledge.” Examples of OBI include the terms “role model,” “deconstruction,” and “self-fulfilling prophecy”.

TE literature has been derived both from the discipline dedicated to the study of teaching (education) and sub-disciplines of teaching, education, and training within each discipline (e.g., accounting education, medical education, computer science education). Our results reflect the mix of education and sub-disciplines contributing to the literature on TE. One possibility is that TE is rarely defined as a construct because it has been “obliterated” by its “incorporation” in all disciplines. Regardless of the reason, the failure to define TE could, in part, be responsible for the lack of a universally accepted measurement for TE (Nielsen, 2014). Similar to TE, as summarized in Table 1, we found that constructs related to TE also had no true definition.

The lack of defining TE and the inconsistency in measuring it may have been detrimental to TE research and thus have a negative effect on teaching, education, and training practices and this interference could impede TE research. Separating the two foci into separate dimensions of TE is consistent with how the measurements are presented in the literature. Sánchez-Cabrero et al. (2021) found the most central factor in teaching quality was related to teachers and their training while student-focused measures are commonly used by universities to evaluate teaching quality (Chau and Vien, 2020). Therefore, we propose pairing an appropriate TE definition with each of the student-focused (results-oriented) or educator-focused (process-oriented) dimensions.

We allocate appropriate definitions to their TE dimension in Table 1. Cohen (1981, p.281) shares, “[TE can be] operationalized as the amount students learn in a particular course.” While McKeachie’s (1979) definition of TE, as used by Galbraith et al. (2012), may appear similar to Cohen’s, it has one distinction: McKeachie operationalizes TE as the facilitation of student achievement of educational goals, whereas Cohen quantifies effective teaching as an amount. This distinction suggests that measuring how much a student learns is best supported by objective measurement of TE, while the facilitation students receive towards that objective achievement may be best assessed via SETE.

When self-assessment, administrator evaluation, and peer review are used to measure TE, it is often done in combination with other TE measures [e.g., Cinnamon et al. (2021) who used objective data (observation) and self-assessment to measure the TE of a new professional development intervention], and is consistent with Reynolds et al. (2014, p.214), “teaching effectiveness is seen as a multi-dimensional construct and a variable factor rather than a universal ‘given’” and Berk (2005), p.49, “unified conceptualization of teaching effectiveness, whereby evidence is collected from a variety of sources to define the construct and to make decisions about its attainment”. We found McKeachie’s (1979) TE definition also fits the educator-focused dimension of TE.

We recommend pairing TE dimension with appropriate TE measurements. Specifically, the student-focused dimension should be paired with student-focused measurements (SETE⁵ and objective), while the educator-focused dimension may be definition should be paired with educator-focused TE measurements (peer review, administrator evaluation, and self-reflection). We observed in Table 1 that SETE has been used to evaluate TE and teacher effectiveness exclusively in higher education, with no instances in primary or secondary education contexts. Similarly, Wei et al. (2023) found that one observation (out of 54 articles reviewed) used SETE to measure teacher effectiveness in primary and secondary education. As such, we recommend removing the SETE measure when evaluating TE in primary or secondary education contexts. Furthermore, we qualify the use of self-assessment and peer review as appropriate to be used in combination with at least one other TE measurement method, and suggest SETE should be used in conjunction with at least one other method when evaluating educator performance. Thus, by pairing the specific dimensions of the TE construct, we provide student-and-educator-focused TE constructs in line with MacKenzie’s (2003) definition of a good construction.

The SETE measure has been problematic throughout the literature. Student evaluations may be gender-biased (Boring et al., 2016; Boring, 2017; Fan et al., 2019), culture-biased (Fan et al., 2019), impacted by “illusions of learning” (e.g., experiences that minimize effort, increase the appearance of fluency, engagement, and enthusiasm) (Carpenter et al., 2020), and impacted by grade expectations (Boring et al., 2016). In their paper titled Availability of cookies during an academic course session affects evaluation of teaching (Hessler et al., 2018) found the availability of cookies during the student’s evaluation of TE to have a significant effect on course evaluation. Despite these drawbacks and research suggesting institutions abandon SETEs (Uttl et al., 2017), SETEs are still used to measure TE. As such, we recommend when using SETE to evaluate educator performance [e.g., for tenure and promotion (Staiger and Rockoff, 2010)], it should be paired with at least one other TE measurement.

To ensure the quality of TE research and contributions to the field, TE must be accurately defined and appropriately measured. Pairing an appropriate definition and measurements for each of the student-and-educator-focused dimensions allow for appropriate TE constructs and enhance the quality of TE research. Such constructs effectively address the question, How is teaching effectiveness conceptualized? And contribute to the academy, educators, administrators, and institutions such that our collective efforts may improve the research impacting students, teachers, and administrators and related organizations.

Our research makes several contributions to the literature. First, we recommend the strongest TE definition by McKeachie (1979, p.385), “we take teaching effectiveness to be the degree to which one has facilitated student achievement of educational goal.” Next, we separate TE into two dimensions: student-focused (outcome) and educator-focused (input) and find that McKeachie’s (1979) TE definition speaks to both educator and student foci, thus recommend their TE definition applies to both student-focused (outcome) and educator-focused (input) dimensions. Lastly, we identify and separate the literature into five types of measurements of TE, SETE, objective, peer review, administrative evaluation, and self-reflection, and sort them as either student- or educator-focused measures of TE. We find that when conceptualizing TE, student-focused measurements (SETE⁵ and objective) should be paired with student-focused definition of TE, and educator-focused measurements (peer review, administrator evaluation, and self-reflection) paired with our educator-focused definition of TE, as summarized in Table 2.

Through our paper we pair an existing strong definition of TE with educator-focused (input) and student-focused (outcome) measures of TE. By conceptualizing TE, researchers may contribute to rigorous research and work together to consistently add to the body of knowledge, thus furthering the quality of TE research.

This study has limitations. We excluded publications that were not peer-reviewed journal publications which may have precluded analysis of literature which may have been additive to our discussion. While we did not exclude any publications by discipline, a systematic literature would have been preferable. In this research, we did not test empirically both dimensions of TE. Consequently, noting the relative disparity of TE definitions in higher education, a future avenue of research could be developing a TE definition specific to the higher education context. We did not empirically test both dimensions of TE. Such empirical study may be an interesting avenue for future research.