Homophily Scales (Old version)

Abstract

The Homophily Scales (Old version), developed by James C. McCroskey, Virginia P. Richmond, and John A. Daly (1974, 1975), represent a foundational measurement battery designed to assess perceived interpersonal similarity in human communication contexts. Rooted in sociological and communication theories of source perception, interpersonal attraction, and diffusion of innovations, homophily reflects the degree to which interacting dyads or individuals perceive themselves as being congruent across diverse cognitive, behavioral, structural, and demographic attributes. The measurement system evolved from an exploratory 46-item semantic differential pool into multidimensional instruments, primarily highlighting Attitude Homophily and Background Homophily, alongside dimensions of Morality Homophily and Appearance Homophily identified across distinct investigative phases. The standardized short forms most frequently deployed in instructional, organizational, and media research consist of four 7-point bipolar semantic differential items per subscale, yielding composite scores bounded between 4 and 28 for each distinct domain. Psychometric assessments of the original scales have repeatedly documented high internal consistency, with Cronbach’s alpha coefficients commonly ranging from .75 to .92 for Attitude Homophily and .65 to .84 for Background Homophily. Construct validity is supported by consistent factorial stability across orthogonal and oblique rotations, strong convergent correlations with measures of source credibility (competence, character, composure, sociability, and extroversion) and interpersonal attraction (social, physical, and task attraction), as well as robust predictive utility regarding communication satisfaction, instructional immediacy, and persuasive compliance. This article offers an exhaustive psychometric examination of the initial 1974–1975 homophily formulation, reviewing its historical development, theoretical foundations, psychometric indicators, scoring mechanisms, and subsequent methodological refinements.

Keywords

Homophily, Perceived Similarity, Attitude Homophily, Background Homophily, James C. McCroskey, Interpersonal Communication, Semantic Differential, Source Credibility, Communication Measurement, Social Attraction

Authors

The Homophily Scales were formulated and psychometrically standardized by three prominent scholars in human communication studies:

• James C. McCroskey, Ed.D. (1938–2012): A pioneering scholar in communication apprehension, instructional communication, and measurement methodology. Dr. McCroskey served as Professor and Chair of the Department of Communication Studies at West Virginia University and held faculty positions at Michigan State University, Old Dominion University, and the University of Hawaii. Across his prolific career, he authored over 300 academic articles and more than 40 books, fundamentally establishing modern instrumentation for communication psychology.
• Virginia P. Richmond, Ph.D.: Professor Emerita of Communication Studies at West Virginia University and later Professor at the University of Alabama at Birmingham. Dr. Richmond is a past President of the Eastern Communication Association (ECA) and an internationally recognized researcher in instructional communication, organizational leadership, nonverbal behavior, and power dynamics.
• John A. Daly, Ph.D.: The Liddell Centennial Professor of Communication, Regents Peace Professor, and University Distinguished Teaching Professor at the University of Texas at Austin. Dr. Daly has contributed seminal research across interpersonal communication, personality correlates of message production, organizational influence, and social skill assessment.

Purpose

The primary purpose of the Homophily Scales (Old version) is to provide empirical researchers and applied social scientists with a standardized, psychometrically rigorous metric for quantifying an individual’s subjective appraisal of interpersonal similarity toward a target source. The phrase “birds of a feather flock together” captures an intuitive social dynamic, yet prior to the 1970s, empirical communication scholarship lacked unified, validated scales capable of isolating whether perceived congruence operated as a unitary phenomenon or as a multidimensional construct spanning beliefs, appearance, socioeconomic standing, and behavioral mannerisms.

In applied and empirical contexts, the instrument serves several pivotal purposes:

• Investigating Persuasive and Source Impact: In speech communication and social influence paradigms, the scales quantify the degree to which receiver perceptions of similarity moderate or mediate source credibility (ethos), attitude change, and message compliance.
• Instructional Communication Diagnostics: Educational researchers utilize the scale to evaluate student-teacher congruence. Documented evidence indicates that perceived homophily between instructors and students substantially predicts affective learning, nonverbal immediacy reciprocation, reduced student resistance, and heightened cognitive engagement.
• Organizational and Dyadic Dynamics: In workplace environments, the instrument assists in analyzing supervisor-subordinate relationships, peer mentoring, negotiation processes, and team cohesion. It operationalizes whether ideological alignment (Attitude Homophily) compensates for structural discrepancies (Background Homophily).
• Mass Media and Health Campaigns: Health communicators and media sociologists deploy the scales to ascertain audience identification with change agents, health spokespersons, or demographic targets in diffusion campaigns, confirming Everett Rogers’ propositions regarding innovative adoption rates.

The conceptual rationale advanced by McCroskey, Richmond, and Daly posits that subjective perception, rather than objective demographic matching, serves as the proximal driver of human communication behavior. Even when objective differences in social status or cultural heritage exist, the communicative outcome is governed predominantly by how similar the receiver perceives the source to be.

Psychological Construct

The construct of perceived homophily denotes the extent to which an individual receiver cognitively construes another target person as being similar to oneself across relevant experiential, psychological, and sociological parameters. Rather than treating similarity as a monolithic variable, the initial investigative models explored by McCroskey et al. (1974, 1975) established that perceived homophily functions as a multifaceted social-cognitive schema. While their foundational 46-item exploratory factor analyses intermittently suggested four distinct components—Attitude, Background, Morality, and Appearance—the definitive operational version crystallized into two primary, empirically robust subscales:

1. Attitude Homophily

Attitude homophily represents the subjective conviction that the source possesses intellectual orientations, values, cognitive processing frameworks, and daily behavioral choices congruent with one’s own. It assesses whether the target “thinks like me,” “behaves like me,” and exhibits an overall worldview aligned with the perceiver. Unlike simple agreement on an isolated political or moral issue, attitude homophily reflects an overarching perceived kinship in how the source interprets reality, processes problems, and reacts affectively to environmental stimuli. It functions as the strongest affective driver within interpersonal attraction frameworks, heavily governing mutual liking, perceived empathy, and interpersonal trust.

2. Background Homophily

Background homophily captures the degree of perceived equivalence regarding social stratification, economic circumstances, cultural origins, status indicators, and life experiences. Items measuring this dimension evaluate whether the target source shares an equivalent socioeconomic footing, occupational level, educational background, or cultural milieu. While objective sociological indices (e.g., income brackets, census categories) can classify demographic matching, perceived background homophily evaluates the subjective psychological reality of whether the perceiver categorizes the target as a member of their own socioeconomic or cultural group (in-group) or as belonging to a distant social stratum (out-group).

3. Ancillary Facets: Appearance and Morality

Across the broader initial 46-item semantic differential matrix, researchers also uncovered distinct clusters reflecting Appearance Homophily (evaluating visual, sartorial, somatic, and aesthetic congruence, such as grooming, clothing, posture, and facial characteristics) and Morality Homophily (evaluating congruence in ethical convictions, moral standards, and religious affiliations). Although subsequent abbreviated scales frequently streamlined testing to focus on Attitude and Background dimensions, the psychological construct encompasses this rich totality of behavioral, physical, ideological, and socio-demographic congruence.

Theoretical Framework

The development of the Homophily Scales relies on the synthesis of several major theoretical paradigms in sociology, social psychology, and communication science:

Lazarsfeld and Merton’s Sociological Homophily

The term homophily was formally introduced into modern sociological theory by Paul F. Lazarsfeld and Robert K. Merton (1954), who differentiated between status homophily (similarity based on formal, ascribed, or acquired social status such as race, socioeconomic position, and age) and value homophily (similarity based on internalized values, attitudes, and personal beliefs). McCroskey, Richmond, and Daly translated this sociological taxonomy directly into communication psychology, recognizing that status homophily corresponds systematically to Background Homophily, whereas value homophily corresponds to Attitude Homophily.

Byrne’s Similarity-Attraction Paradigm

In social psychology, Donn Byrne’s (1971) similarity-attraction effect provided the behavioral mechanism explaining why perceived homophily enhances interpersonal communication. Byrne demonstrated experimentally that individuals experience consensual validation when interacting with congruent others. When an interlocutor manifests similar cognitive positions, it reinforces the receiver’s personal construction of reality, generating positive reinforcement, lowering interpersonal anxiety, and fostering affective attraction. McCroskey and colleagues situated the measurement of homophily as a necessary prerequisite to assessing how interpersonal attraction mediates message reception.

Rogers’ Diffusion of Innovations

Everett M. Rogers (1962, 1973) posited that effective communication transfer occurs most effortlessly between homophilous individuals. In contrast, heterophilous dyads—characterized by divergent educational backgrounds, beliefs, or socioeconomic circumstances—face substantial perceptual and semantic barriers. Rogers underscored an inherent paradox in social change: while homophilous communication is exceptionally smooth, the introduction of innovative ideas often requires interaction across heterophilous boundaries. McCroskey et al. recognized the critical need for a reliable operational measure of this construct to systematically evaluate diffusion processes within small groups, instructional settings, and broad media networks.

Validity

Extensive empirical investigation across several decades has validated the psychometric robustness of the Homophily Scales (Old version):

Construct and Factorial Validity

During initial scale validation, McCroskey, Richmond, and Daly (1975) administered the comprehensive 46-item semantic differential matrix to diverse participant samples responding to diverse target personas, including peers, superiors, teachers, political figures, and conversational partners. Factor analyses (principal components with varimax and promax rotations) repeatedly isolated clean, interpretable factors corresponding to the hypothesized dimensions. The four-item operational subscales for Attitude and Background homophily consistently demonstrated high primary factor loadings (> .60) and minimal cross-loadings (< .30), demonstrating sharp structural integrity.

Convergent and Concurrent Validity

The scales demonstrate substantial convergent validity through their theoretical associations with interpersonal attraction and source credibility. In studies evaluating student-teacher interactions and interpersonal dyads (e.g., Rocca & McCroskey, 1999; McCroskey, McCroskey, & Richmond, 2006):

• Attitude Homophily strongly correlates with Social Attraction (typically r = .55 to .72) and Task Attraction (r = .40 to .60) on the Interpersonal Attraction Scale (McCroskey & McCain, 1974).
• Attitude Homophily correlates positively with student evaluations of instructor Sociability, Competence, and Character, as well as perceptions of teacher Nonverbal Immediacy (r values typically exceeding .45).
• Background Homophily displays moderate, statistically significant correlations with socioeconomic index matching, demonstrating concurrent validity with objective demographic measures while capturing perceptual variance.

Discriminant and Predictive Validity

Discriminant validity is supported by the empirical separation between homophily subdimensions. Attitude Homophily and Background Homophily regularly correlate with each other at modest to moderate levels (r values typically ranging between .20 and .45), demonstrating that respondents clearly distinguish between someone having an equivalent social background and someone thinking or behaving like them. Furthermore, predictive validity has been verified across diverse settings: ratings of attitude homophily reliably predict longitudinal relational development, conversational satisfaction, willingness to communicate, and compliance with behavioral recommendations.

Reliability

The internal consistency of the Homophily Scales has been documented across dozens of peer-reviewed communication and psychological studies:

• Attitude Homophily Subscale: Across varied research populations (undergraduate students, corporate employees, healthcare recipients), the 4-item Attitude Homophily scale consistently yields Cronbach’s alpha coefficients between .78 and .92. In McCroskey, Richmond, and Daly’s (1975) foundational investigations, internal consistency across target conditions routinely met or exceeded .85.
• Background Homophily Subscale: The 4-item Background Homophily scale demonstrates acceptable to solid internal consistency, with alpha coefficients typically falling between .68 and .84. The slightly broader dispersion of scores reflects the multidimensional nature of social background (economic, cultural, status, and class indicators).
• Test-Retest Stability: In longitudinal dyadic research monitoring relational progress over 2-week to 6-week intervals without significant relational disruptions, test-retest reliability coefficients have been found to exceed .75, indicating that perceptual schemas of homophily remain reasonably stable over moderate durations in the absence of new disconfirming information.

Factor Analysis

The factor structure of the original homophily instrument was delineated using exploratory factor analytic (EFA) procedures on a 46-item semantic differential inventory. In the foundational study by McCroskey, Richmond, and Daly (1975), several hundred participants rated familiar targets across a continuum of relational closeness:

Exploratory Factor Solutions

Initial principal components factor analysis with orthogonal (Varimax) rotation extracted four primary factors accounting for substantial cumulative variance (> 50%):

• Factor I (Attitude): Characterized by high loadings from items assessing perceptual, cognitive, and behavioral congruence (“Is like me / Is unlike me,” “Is different from me / Is similar to me,” “Thinks like me / Does not think like me,” “Doesn’t behave like me / Behaves like me”). Primary loadings ranged from .68 to .82, with negligible cross-loadings onto status or demographic factors.
• Factor II (Background): Anchored by items addressing external social standing, class, culture, and wealth (“Has status like mine / Has status different from mine,” “Is from a different social class / Is from the same social class,” “Is culturally different / Is culturally similar,” “Has an economic situation like mine / Does not have an economic situation like mine”). Primary factor loadings ranged between .58 and .76.
• Factor III (Appearance) & Factor IV (Morality): Evident in the expanded 46-item research instrument, these factors isolated items assessing somatic, sartorial, and stylistic similarity (clothing, hair, posture, weight, size) and ethical/normative congruence (morals, values, sexual attitudes, religious affiliation).

Structural Clarifications and Subsequent CFA

In subsequent confirmatory factor analytic (CFA) investigations examining the measurement properties of homophily (e.g., McCroskey, McCroskey, & Richmond, 2006), the brief four-item Attitude and four-item Background scales showed adequate construct differentiation. While some items in the broader 46-item research instrument exhibited cross-loadings across distinct communicative contexts, the two core 4-item operational scales retained robust unidimensionality within their respective constructs, showing acceptable goodness-of-fit indices (CFI > .94, RMSEA < .07) when specified as two correlated latent factors.

Instrument / Measurement Tool

The original measurement system encompasses two distinct formats: the core Operational Homophily Scales (consisting of two 4-item subscales frequently utilized in empirical questionnaires) and the comprehensive 46-Item Research Instrument (utilized in detailed psychometric evaluations).

Format and Structure

• Target Specification: A prompt at the top designates the target persona: (Name of person to whom participant is asked to respond).
• Response Continuum: 7-point bipolar semantic differential scale.
• Scale Point Anchors: Numbers 1 and 7 represent “a very strong feeling”; numbers 2 and 6 indicate “a strong feeling”; numbers 3 and 5 represent “a fairly weak feeling”; number 4 designates an “unsure or undecided” neutral mid-point.
• Subscale Item Counts: Core Attitude Homophily (4 items); Core Background Homophily (4 items); Comprehensive Research Instrument pool (46 items).

Scoring Instructions

Scoring proceeds separately for each subscale. Because polarity is counterbalanced across items to attenuate acquiescence response set bias, researchers must ensure proper reverse coding prior to calculation so that higher numbers systematically indicate higher perceived homophily:

• Attitude Homophily Scoring:
  • Item 1 (Is like me = 7 to Is unlike me = 1): Directly coded.
  • Item 2 (Is different from me = 1 to Is similar to me = 7): Directly coded.
  • Item 3 (Thinks like me = 7 to Does not think like me = 1): Directly coded.
  • Item 4 (Doesn’t behave like me = 1 to Behaves like me = 7): Directly coded.
  • Calculation: Sum the values of items 1, 2, 3, and 4. Total scores range from 4 to 28. Higher scores reflect greater perceived attitude similarity.
• Background Homophily Scoring:
  • Item 1 (Has status like mine = 7 to Has status different from mine = 1): Directly coded.
  • Item 2 (Is from a different social class = 1 to Is from the same social class = 7): Directly coded.
  • Item 3 (Is culturally different = 1 to Is culturally similar = 7): Directly coded.
  • Item 4 (Has an economic situation like mine = 7 to Does not have an economic situation like mine = 1): Directly coded.
  • Calculation: Sum the values of items 1, 2, 3, and 4. Total scores range from 4 to 28. Higher scores reflect greater perceived background similarity.

Permissions & Fee and Test Year

The core developmental articles for the Homophily Scales were presented at the 1974 International Communication Association (ICA) convention in New Orleans and formally published in Human Communication Research in 1975. Under the academic measurement policy established by Dr. James C. McCroskey and maintained by the Department of Communication Studies at West Virginia University, these measurement instruments are placed in the public domain for research and scholarly instructional use without fee. Researchers, clinicians, and educators may utilize, adapt, and reproduce the scales provided that appropriate scholarly attribution is accorded to the original authors (McCroskey, Richmond, & Daly, 1975). Commercial applications, software packaging, or for-profit organizational licensing typically require formal clearance through the estate of James C. McCroskey or the original publishing copyright holders.

References

• Byrne, D. (1971). The Attraction Paradigm. Academic Press.
• Lazarsfeld, P. F., & Merton, R. K. (1954). Friendship as a social process: A substantive and methodological analysis. In M. Berger, T. Abel, & C. H. Page (Eds.), Freedom and Control in Modern Society (pp. 18–66). Van Nostrand.
• McCroskey, J. C., & McCain, T. A. (1974). The measurement of interpersonal attraction. Speech Monographs, 41(3), 261–266. https://doi.org/10.1080/03637757409375845
• McCroskey, J. C., Richmond, V. P., & Daly, J. A. (1974). Toward the measurement of perceived homophily in interpersonal communication. Paper presented at the Annual Meeting of the International Communication Association, New Orleans, LA, April 17–20. https://files.eric.ed.gov/fulltext/ED094433.pdf
• McCroskey, J. C., Richmond, V. P., & Daly, J. A. (1975). The development of a measure of perceived homophily in interpersonal communication. Human Communication Research, 1(4), 323–332. https://doi.org/10.1111/j.1468-2958.1975.tb00281.x
• McCroskey, L. L., McCroskey, J. C., & Richmond, V. P. (2006). Analysis and improvement of the measurement of interpersonal attraction and homophily. Communication Quarterly, 54(1), 1–31. https://doi.org/10.1080/01463370500270310
• Rocca, K. A., & McCroskey, J. C. (1999). The interrelationship of student ratings of instructors’ immediacy, verbal aggressiveness, homophily, and interpersonal attraction. Communication Education, 48(4), 308–316. https://doi.org/10.1080/03634529909379181
• Rogers, E. M. (1962). Diffusion of Innovations. Free Press.
• Rogers, E. M., & Bhowmik, D. K. (1970). Homophily-heterophily: Relational concepts for communication research. Public Opinion Quarterly, 34(4), 523–538. https://doi.org/10.1086/267838

13. Items of the Scale (Questionnaire)