Measure proliferation as a compounding threat to our future

Bowling et al.’s focal article, Read my lips: No new constructs! Construct proliferation as a threat to the future of I-O psychology (Reference Bowling, Sessa, Shaffer and Banks2026), offers a timely and compelling diagnosis of a problem that has quietly intensified within industrial–organizational (I-O) psychology. As the authors document, construct proliferation, the introduction of ostensibly distinct constructs that are conceptually or empirically redundant with existing ones, has become increasingly prevalent, with the number of recognized constructs in I-O psychology more than doubling since 2013 (Anvari et al., Reference Anvari, Alsalti, Oehler, Marion, Hussey, Elson and Arslan2025). The authors persuasively argue that although construct proliferation may occasionally serve useful purposes, it is “usually bad,” largely inevitable, and frequently incentivized by prevailing academic reward structures (Bowling et al., Reference Bowling, Sessa, Shaffer and Banks2026, p. 2). Most concerning, construct proliferation undermines parsimony, impedes cumulative knowledge, and complicates the translation of research findings into practice.

We largely concur with this assessment and applaud the focal article’s call for an increased scrutiny of the introduction of new constructs. Nonetheless, we believe it is important to maintain a distinction between construct and method (e.g., Arthur & Villado, Reference Arthur and Villado2008). In this commentary, we extend this conversation by arguing that measure proliferation, the proliferation of multiple, often redundant, but potentially nonequivalent instruments intended to assess the same construct, poses an additional and compounding threat to theory development, empirical synthesis, and applied utility. Whereas construct proliferation concerns the multiplication of labels and conceptual definitions for similar phenomena, measure proliferation concerns the unchecked multiplication of scales and instruments used to operationalize a single construct. Although these problems are analytically distinct, they are deeply intertwined in practice, and addressing construct proliferation without simultaneously confronting measure proliferation risks leaving a major source of fragmentation intact.

Defining measure proliferation

Measure proliferation occurs when a single psychological construct is assessed using numerous overlapping but potentially nonequivalent instruments, each purportedly offering incremental improvement, novelty, or specialization, but without clear evidence that such proliferation meaningfully enhances construct validity or practical usefulness. Unlike construct proliferation, which often involves rhetorical and definitional differentiation, measure proliferation frequently arises even when researchers agree on the construct. In other words, consensus at the conceptual level does not guarantee consensus at the operational level.

Importantly, measure proliferation is not inherently problematic. The availability of multiple instruments can be beneficial when measures are empirically demonstrated to be equivalent (e.g., parallel forms) or when substantive theoretical advances necessitate the development of new instruments aligned with updated conceptualizations. Problems arise, however, when multiple measures of ostensibly the same construct coexist without clear evidence of conceptual alignment or empirical equivalence, creating ambiguity about what is being measured and undermining cumulative knowledge building.

Thus, as with construct proliferation, some degree of proliferation is expected (and even desirable) during early phases of construct development. New measures may be introduced to address psychometric limitations of earlier instruments, improve subgroup invariance, reduce respondent burden, or adapt assessment for certain research designs such as experience sampling methodology (ESM), which require shorter scales. However, when the field fails to converge on an established, evidence-based set of preferred instruments after such exploratory phases, measure proliferation becomes a persistent source of noise rather than signal.

Why measure proliferation occurs

Many of the same forces that Bowling et al. (Reference Bowling, Sessa, Shaffer and Banks2026) identify as drivers of construct proliferation also fuel measure proliferation. First, researchers may genuinely believe that existing measures are psychometrically insufficient. Poor factor loadings, limited reliability, or lack of measurement invariance may motivate the development of revised or alternative scales. This is the soundest reason to create a new measure when one or more already exist. Second, practical constraints like those that exist in longitudinal, ESM, or applied contexts often lead to the creation of abbreviated forms rather than the systematic evaluation of existing ones. Creating an abbreviated form is not an inherently bad practice. However, reporting of adaptations (such as the shortening of scales) is often insufficient (Heggestad et al., Reference Heggestad, Scheaf, Banks, Hausfeld, Tonidandel and Williams2019), and these abbreviations are a key area where proliferation can quickly occur as researchers continue to significant adapt existing measure rather than converging on a single abbreviated form.

Third, historical drift plays a role. As constructs evolve conceptually, existing measures may be perceived as outdated, even when they remain psychometrically sound. As Bowling et al. (Reference Bowling, Sessa, Shaffer and Banks2026) note, cascading adaptations of validated scales can result in items that gradually become less representative of the original construct (Heggestad et al., Reference Heggestad, Scheaf, Banks, Hausfeld, Tonidandel and Williams2019). Fourth, as with construct proliferation, researchers may be unaware that a suitable measure already exists, particularly when relevant work spans disciplinary or subdisciplinary boundaries.

Finally, and perhaps most consequentially, the academic incentive structure is not particularly conducive to the in-depth and systematic development of new measures. We would argue that the field’s increasing emphasis on new theory development attenuates the publication of papers devoted to the systematic development and evaluation of measures. Often measures are presented in the context of concurrent empirical studies and information pertaining to scale development and evaluation is superficial at best. These measures are then subsequently used by other researchers and cited as established operationalizations of the construct of interest. Even worse, these measures are often “adapted” in a way that further limits validity (Heggestad et al., Reference Heggestad, Scheaf, Banks, Hausfeld, Tonidandel and Williams2019). In this sense, measure proliferation is not simply an individual-level problem but a systemic one, reinforced by the same institutional forces that Bowling et al. (Reference Bowling, Sessa, Shaffer and Banks2026) identify as incentivizing construct proliferation.

Why measure proliferation is also usually bad

Measure proliferation compounds the very problems that construct proliferation creates. When multiple constructs with similar labels exist, researchers and practitioners must already navigate conceptual ambiguity. When each of those constructs is assessed using multiple, overlapping, and potentially nonequivalent measures, the burden of interpretation grows exponentially. Researchers conducting reviews or meta-analyses face a fundamental question: Are studies using different instruments truly assessing the same construct, or are apparent differences in findings driven by idiosyncrasies of measurement?

These issues are particularly acute for cumulative science. Without convergence at the measurement level, it becomes difficult to determine whether inconsistent findings reflect substantive theoretical differences or merely differences in operationalization. As Bowling et al. observe, construct proliferation “makes it more difficult for I-O researchers to work together, to build off the work of others, and to accumulate knowledge across studies” (Bowling et al., Reference Bowling, Sessa, Shaffer and Banks2026, p. 15). Measure proliferation magnifies this problem by obscuring whether researchers are, in fact, studying the same phenomenon.

For practitioners, the consequences are equally troubling. Practitioners often lack the time, resources, or methodological expertise to evaluate competing measures. Faced with an abundance of options, they may rely on heuristics such as recency or citation counts to select instruments. Yet the most recent or most cited measure is not necessarily the most valid or comprehensive. As the focal article notes, construct proliferation already makes it difficult for practitioners to interpret and implement research-based recommendations, pushing stakeholders toward clearer (but less evidence-based) alternatives (pp. 15–16). Measure proliferation further erodes the accessibility and credibility of I-O psychology by making even well-established constructs appear unsettled and fragmented.

An illustrative example: Measurement of conscientiousness

The scope of measure proliferation becomes especially apparent when examining long-standing, ostensibly well-defined constructs. For example, we conducted a targeted examination of measures of conscientiousness in primary articles published over the past 5 years in three leading journals (Journal of Applied Psychology, Journal of Management, and Academy of Management Journal). Using a conservative set of search terms related to measurement (e.g., scale, questionnaire, inventory), we identified 12 unique conscientiousness scales used within this narrow slice of the literature.

This finding is striking precisely because conscientiousness is among the most theoretically mature and empirically validated constructs in our field. If such proliferation exists for a core Big Five trait, it is reasonable to expect similar or even greater proliferation for other established constructs. Importantly, the presence of multiple conscientiousness measures does not, by itself, indicate conceptual disagreement. Rather, it reflects a lack of systematic effort to evaluate competing instruments and establish consensus with respect to how an important construct should be assessed in the literature.

Diagnosing versus solving the problem

Some research has begun to diagnose the scope and consequences of measure proliferation. Martinko et al. (Reference Martinko, Mackey, Moss, Harvey, McAllister and Brees2018), for example, demonstrated substantial overlap among leadership evaluation measures and showed that much of this shared variance could be explained by subordinate affect. Their work illustrates how careful empirical analysis can reveal redundancy at both the construct and measure levels. However, diagnosis alone is only the first step. Identifying overlap without articulating clear recommendations leaves researchers and practitioners uncertain about how to proceed.

To meaningfully address measure proliferation for core constructs, the field must move beyond cataloging redundancy and toward selective consolidation. Once evidence of measure proliferation is diagnosed, researchers should explicitly evaluate competing instruments and determine which measures best represent the construct’s full content space, demonstrate robust psychometric performance, and generalize across contexts. Instruments that consistently underperform, or that offer no clear advantage over alternatives, should be retired.

Toward a coordinated response

Addressing measure proliferation requires coordinated action across multiple levels of the field. First, novel measures of existing constructs should be approached with skepticism comparable to that advocated by Bowling et al. (Reference Bowling, Sessa, Shaffer and Banks2026) for new constructs. Authors should clearly articulate why existing measures are insufficient and provide direct evidence that the proposed instrument offers meaningful improvement. Reviewers and editors play a critical gatekeeping role here by requiring such justification.

Second, journals must create space for measure development, evaluation, and consolidation work. Just as Bowling et al. (Reference Bowling, Sessa, Shaffer and Banks2026) call for special issues focused on construct proliferation, similar outlets are needed for systematic evaluations of competing measures and/or work that tackles these issues in tandem. Such work is unlikely to be incentivized under current norms unless journals explicitly signal its value.

Third, the field would benefit from a centralized, accessible repository of recommended measures for core I-O constructs. Although prior efforts (e.g., the International Personality Item Pool [IPIP]), have sought to increase transparency and accessibility by housing large collections of items and scales, these resources stop short of providing evaluative guidance. As a result, researchers are still left to navigate many competing measures for the same construct with little clarity or guidance regarding which instruments are most strongly supported by validation evidence. A centralized repository could address this gap by documenting the history of measure development, synthesizing comparative evidence across scales, and offering clear, evidence-based recommendations (e.g., one well-supported long form and one short form) for each core construct. Importantly, such a resource would not freeze measurement development but would make updates transparent and cumulative rather than ad hoc, allowing researchers to see not only which measures are currently recommended but also why alternative instruments are less strongly supported based on existing evidence.