Amir Goldberg

Stanford Graduate School of Business


Aligning Differences: Discursive Diversity and Team Performance

with Katharina Lix, Sameer B. Srivastava and Melissa A. Valentine

Management Science, forthcoming

How does cognitive diversity in a group affect its performance? Prior research suggests that group cognitive diversity poses a performance tradeoff: diverse groups excel at creativity and innovation but struggle to take coordinated action. Building on the insight that group cognition is not static but is instead dynamically and interactively produced, we develop a novel conceptualization of group cognitive diversity--discursive diversity, or the degree to which the meanings conveyed by group members in a given set of interactions diverge from one another. We propose that high-performing teams are ones that have a collective capacity to modulate shared cognition to match changing task requirements: they exhibit higher discursive diversity when engaged in ideational tasks and lower discursive diversity when performing coordination tasks. We further argue that teams exhibiting congruent modulation--that is, those with low group-level variance in members' within-person semantic shifts to changing task requirements--are more likely to experience success than teams characterized by incongruent modulation. Using the tools of computational linguistics to derive a measure of discursive diversity and drawing on a novel longitudinal data set of intragroup electronic communications and performance outcomes for 117 remote software development teams on an online platform (, we find support for our theory. Our findings suggest that the performance tradeoff of group cognitive diversity is not inescapable: groups can navigate it by aligning their levels of discursive diversity to match their task requirements and by having members stay aligned with one another as they make these adjustments.

Duality in Diversity: Cultural Heterogeneity, Language, and Firm Performance

with Matthew Corritore and Sameer B. Srivastava

Administrative Science Quarterly Vol. 65 No. 2 (2020))

How does cultural heterogeneity in an organization relate to its underlying capacity for execution and innovation? Existing literature often understands cultural diversity as presenting a trade-off between task coordination and creative problem-solving. This work assumes that diversity arises primarily through cultural differences between individuals. In contrast, we propose that diversity can also exist within persons such that cultural heterogeneity can be unpacked into two distinct forms: interpersonal and intrapersonal. We argue that the former tends to undermine coordination and portends worsening firm profitability, while the latter facilitates creativity and supports greater patenting success and more positive market valuations. To evaluate these propositions, we use unsupervised learning to identify cultural content in employee reviews of nearly 500 publicly traded firms on a leading company review website and then develop novel, time-varying measures of cultural heterogeneity. Our empirical results lend support for our two core propositions, demonstrating that a diversity of cultural beliefs in an organization does not necessarily impose a trade-off between operational efficiency and creativity.

Beyond "Social Contagion": Associative Diffusion and the Emergence of Cultural Variation

with Sarah K. Stein

American Sociological Review Vol. 83 No. 5 (2018)

Source Code

Network models of diffusion predominantly think about cultural variation as a product of social contagion. But culture does not spread like a virus. In this paper, we propose an alternative explanation which we refer to as associative diffusion. Drawing on two insights from research in cognition--that meaning inheres in cognitive associations between concepts, and that such perceived associations constrain people's actions--we propose a model wherein, rather than beliefs or behaviors per-se, the things being transmitted between individuals are perceptions about what beliefs or behaviors are compatible with one another. Conventional contagion models require an assumption of network segregation to explain cultural variation. In contrast, we demonstrate that the endogenous emergence of cultural differentiation can be entirely attributable to social cognition and does not necessitate a clustered social network or a preexisting division into groups. Moreover, we show that prevailing assumptions about the effects of network topology do not hold when diffusion is associative.

Enculturation Trajectories: Language, Cultural Adaptation, and Individual Outcomes in Organizations

with Sameer B. Srivastava, V. Govind Manian and Christopher Potts

Management Science Vol. 64 No. 3 (2018)

How do people adapt to organizational culture and what are the consequences for their outcomes in the organization? These fundamental questions about culture have previously been examined using self-report measures, which are subject to reporting bias, rely on coarse cultural categories defined by researchers, and provide only static snapshots of cultural fit. In contrast, we develop an interactional language use model that overcomes these limitations and opens new avenues for theoretical development about the dynamics of organizational culture. We trace the enculturation trajectories of employees in a mid-sized technology firm based on analyses of 10.24 million internal emails. Our language-based model of changing cultural fit: (1) predicts individual attainment; (2) reveals distinct patterns of adaptation for employees who exit voluntarily, exit involuntarily, and remain employed; (3) demonstrates that rapid early cultural adaptation reduces the risk of involuntary, but not voluntary, exit; and (4) finds that a decline in cultural fit for individuals who had successfully enculturated portends voluntary departure.

Searching for Homo Economicus: Variation in Americans' Construals of and Attitudes toward Markets

with Paul DiMaggio

European Journal of Sociology Vol. 59 No. 2 (2018)

Economic sociologists agree that economic rationality is constructed and that morality and economic interests often intersect. Yet we know little about how Americans organize their economic beliefs or assess the morality of markets. To make progress, it is neces­sary to distinguish between how actors construe markets (how they understand and structure their attitudes toward markets) and their normative positions on markets’ prop­er role. Using data from the General Social Survey, we employ Relational Class Analysis to identify three sub­sets of respondents whose members construe economic markets in distinct ways. Compared to the full sample, subsamples display markedly more structure in associations among responses, and between attitudes and sociodem­ographic predict­ors. Support for market solutions is associated with indicators of economic advantage in each subset, but religious and political identities, respectively, predict pro-market views uniquely in subsamples that construe markets through a religious or political lens. Re­sults illustrate the value of distinguishing between construals and positions and examin­ing population heterogeneity in opinion data; identify and explain variations in how Am­ericans understand markets; and illuminate the moral dimension in economic attitudes. Self-interest drives faith in markets, but only when people construe markets in ways consistent with their religious and political faiths.

Alignment at Work: Using Language to Distinguish the Internalization and Self-Regulation Components of Cultural Fit in Organizations

with Gabriel Doyle, Sameer B. Srivastava and Michael C. Frank

Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (2017)

Cultural fit is widely believed to affect the success of individuals and the groups to which they belong. Yet it remains an elusive, poorly measured construct. Recent research draws on computational linguistics to measure cultural fit but overlooks asymmetries in cultural adaptation. By contrast, we develop a directed, dynamic measure of cultural fit based on linguistic alignment, which estimates the influence of one person's word use on another's and distinguishes between two enculturation mechanisms: internalization and self-regulation. We use this measure to trace employees' enculturation trajectories over a large, multi-year corpus of corporate emails and find that patterns of alignment in the first six months of employment are predictive of individuals downstream outcomes, especially involuntary exit. Further predictive analyses suggest referential alignment plays an overlooked role in linguistic alignment.

Culture out of Attitudes: Relationality, Population Heterogeneity and Attitudes toward Science and Religion in the U.S.

with Paul J. DiMaggio, Ramina Sotoudeh and Hana Shepherd

Poetics Vol. 68 (2018)

Attitude data can reveal culture's secrets, but only if analysts acknowledge and transcend two problematic forms of heterogeneity. The first, relational heterogeneity, reflects the fact that the meaning of a response to a survey attitude question emerges from its relation to other attitudes: considered singly, the same response may mean different things to different respondents, depending upon the meanings with which they associate it. The second, population heterogeneity, a common problem in survey analysis, reflects the fact that attitudes may be related to one another in systematically different ways for different respondent subsamples. To overcome these challenges, we must use analytic methods that (a) focus on relations among attitude responses rather than on single responses and (b) partition survey samples into subsets based on patterns emergent from those relations. We use two such approaches, Latent Class Analysis and Relational Class Analysis, to examine Americans' attitudes toward science and religion in the late 20th century, at the onset of a period of acute cultural contention between religious conservatives and secular liberals. Employing an unusually rich data set that enables us to take into account spiritualism (supernatural experience not sanctioned by formal religious institutions), as well as science and religion, we find that both LCA and RCA identify large subsets of respondents for whom science and religion are allied, rather than opposed. Moreover, RCA enables us to examine how the determinants of attitudes toward science, religion, and spiritualism are conditioned upon respondents' construals of the relationships among them. This diversity of opinion among religious Americans and the presence of a previously overlooked religious constituency of science supporters, has important implications for science policy and science advocacy.

What is Cultural Fit? From Cognition to Behavior (and Back)

with Sanaz Mobasseri and Sameer B. Srivastava

Oxford Handbook of Cognitive Sociology

How people fit into social groups is a core topic of investigation across multiple sociological subfields, including education, immigration, and organizations. In this chapter, we synthesize findings from these literatures to develop an overarching framework for conceptualizing and measuring the level of cultural fit and the dynamics of enculturation between individuals and social groups. We distinguish between the cognitive and behavioral dimensions of fitting in, which previous work has tended to either examine in isolation or to conflate. Reviewing the literature through this lens enables us to identify the strengths and limitations of unitary-that is, primarily cognitive or primarily behavioral-approaches to studying culturalfit. In contrast, we develop a theoretical framework that integrates the two perspectives and highlights the value of considering their interplay over time. We then identify promising theoretical pathways that can link the two dimensions of cultural fit. We conclude by discussing the implications of pursuing these conceptual routes for research methods and provide some illustrative examples of such work.

Spillovers inside Conglomerates: Incentives and Capital

with Ran Duchin and Denis Sosyura

Review of Financial Studies Vol. 30 No. 5 (2017)

Using hand-collected data on divisional managers at conglomerates, we find that a change in industry surplus in one division generates large spillovers on managerial payoffs in other divisions of the same firm. These spillovers arise only within the boundaries of a conglomerate but not between standalone firms that match conglomerates' divisions. The intra-firm spillovers increase when conglomerates have excess cash and when managers have more influence over its distribution, but decline in the presence of strong shareholder governance. These spillovers are associated with weaker performance and lower firm value. Our evidence is consistent with simultaneous cross-subsidization via managerial payoffs and capital budgets and suggests that these practices arise in similar firms.

Fitting In or Standing Out? The Tradeoffs of Structural and Cultural Embeddedness

with Sameer B. Srivastava, V. Govind Manian, Will Monroe and Christopher Potts

American Sociological Review Vol. 81 No. 6 (2016)

A recurring theme in sociological research is the tradeoff between fitting in and standing out. Prior work examining this tension has tended to take either a network structural or a cultural perspective. We instead fuse these two traditions to develop a theory of how structural and cultural embeddedness jointly relate to individual attainment within organizations. Given that organizational culture is hard to observe, we develop a novel approach to assessing individuals' cultural fit with their colleagues in an organization based on the language expressed in internal email communications. Drawing on a unique data set that includes a corpus of 10.25 million email messages exchanged over five years among 601 employees in a high-technology firm, we find that network constraint impedes, while cultural fit promotes, individual attainment. More importantly, we find evidence of a tradeoff between the two forms of embeddedness: cultural fit benefits individuals with low network constraint (i.e., brokers), while network constraint promotes attainment for those with low cultural fit.

What Does It Mean to Span Cultural Boundaries? Variety and Atypicality in Cultural Consumption

with Michael T. Hannan and Balázs Kovács

American Sociological Review Vo. 81 No. 2 (2016)

We propose a synthesis of two lines of sociological research on boundary spanning in cultural production and consumption. One, research on cultural omnivorousness, analyzes choice by heterogeneous audiences facing an array of crisp cultural offerings. The other, research on categories in markets, analyzes reactions by homogeneous audiences to objects that vary in the degree to which they conform to categorical codes. We develop a model of heterogeneous audiences evaluating objects that vary in typicality. This allows consideration of orientations on two dimensions of cultural preference: variety and typicality. We propose a novel analytical framework to map consumption behavior in these two dimensions. We argue that one audience type, those who value variety and typicality, are especially resistant to objects that span boundaries. We test this argument in an analysis of two large-scale datasets of reviews of films and restaurants.

Neither Ideologues, nor Agnostics: Alternative Voters' Belief System in an Age of Partisan Politics

with Delia Baldassarri

American Journal of Sociology Vol. 120, No. 1 (2014)

Most research on public opinion assumes that American political views are structured by a belief system with a clearly defined liberal-conservative polarity; however, this is not true of all Americans. In this article we document systematic heterogeneity in the organization of political attitudes and explain its basis in the sociodemographic profile of the respondents. We use Relational Class Analysis (RCA), a network-based method for detecting heterogeneity in collective patterns of opinion, to identify distinctive belief networks, each shared by a different group of respondents. Analyzing ANES data between 1984 and 2004, we identify three groups of American citizens: Ideologues, whose political attitudes strongly align with either liberal or conservative categories; Alternatives, who are instead morally conservative but economically liberal, or vice versa; and Agnostics, who exhibit weak associations among political beliefs. Respondents' sociodemographic profiles, particularly their income, education,and religiosity, lie at the core of the different ways in which they understand politics.

Mapping Shared Understandings Using Relational Class Analysis: The Case of the Cultural Omnivore Reexamined

American Journal of Sociology Vol. 116 No. 5 (2011)

Sociologists often describe culture as a repertoire of shared understandings. But because the meanings that social actors attribute to symbols and actions emerge from the multiple associations they make between them, delineating collectively shared understandings is not a straightforward task. Standard quantitative sociological practice, which relies on the assumptions of what Abbott (1988) calls 'general linear reality', falls short of addressing such complexity in two significant ways: first, it overlooks the multivocality of cultural objects by presupposing that the effects that social attributes have on cultural interpretations are consistent across individuals, and second, it fails to acknowledge that people may have different behaviors or opinions on particular issues, but still agree on the structures of relevance and opposition that make symbols and actions meaningful. In this paper, I introduce a new method - Relational Class Analysis - that simultaneously compares differences between individuals, and within their sets of attitudes, as a means to detect groups with underlying shared understandings of a particular social domain. To demonstrate the utility of this method, I use it to revisit the cultural omnivore thesis by reexamining Americans' attitudes toward musical genres. I find further support for the claim that high-status individuals have replaced cultural snobbism with inclusiveness, but also demonstrate the existence of two competing and systematically overlooked logics of distinction: one that continues to distinguish between high- and low-brow music, the other which distinguishes between traditional and contemporary musical preferences. These findings complicate, and in some ways challenge, contemporary understandings of cultural omnivorousness.

In Defense of Forensic Social Science

Big Data and Society, July-December 2015

Like the navigation tools that freed ancient sailors from the need to stay close to the shoreline - eventually affording the discovery of new worlds - Big Data might open us up to new sociological possibilities by freeing us from the shackles of hypothesis testing. But for that to happen we need forensic social science: the careful compilation of evidence from unstructured digital traces as a means to generate new theories.

Sociology in the Era of Big Data: The Ascent of Forensic Social Science

with Daniel A. McFarland and Kevin Lewis

The American Sociologist Vol. 47 No. 1 (2015)

We propose a synthesis of two lines of sociological research on boundary spanning in cultural production and consumption. One, research on cultural omnivorousness, analyzes choice by heterogeneous audiences facing an array of crisp cultural offerings. The other, research on categories in markets, analyzes reactions by homogeneous audiences to objects that vary in the degree to which they conform to categorical codes. We develop a model of heterogeneous audiences evaluating objects that vary in typicality. This allows consideration of orientations on two dimensions of cultural preference: variety and typicality. We propose a novel analytical framework to map consumption behavior in these two dimensions. We argue that one audience type, those who value variety and typicality, are especially resistant to objects that span boundaries. We test this argument in an analysis of two large-scale datasets of reviews of films and restaurants.

Working Papers

Differences Beyond Identity: Perceived Construal Distance and Interparty Animosity in the United States

with Austin van Loon and Sameer B. Srivastava

What leads people to feel negatively about members of other social groups? While recent work has mostly focused on the strength of group identity, we propose that perceptions of intergroup difference, and the animus they catalyze, are often rooted in beliefs that members of other groups interpret the world in inherently different ways. Drawing on work in the sociology of culture, we define these interpretations---which we refer to as construals---as cognitive associations and develop a theory of multilevel perceived construal distance. We theorize that perceived intergroup construal distance relates to negative outgroup affect, whereas perceptions of construal distance between an outgroup and "most people" relates to outgroup dehumanization. We introduce a new method, the Multilevel Construal Elicitation Task (MCET), to tap respondents' multilevel construals and apply it to politically affiliated Americans' perceptions of the concept "America." In support of our theory, we demonstrate that antipathy toward members of the opposing party is related to people's perceptions of construal distance, above and beyond their strength of party identity or ideology. Our findings contribute to debates on the causes of increased political polarization in the U.S. and have implications for research on identity, social boundaries, and measuring cultural meaning.

Who Sees the Future? A Deep Learning Language Model Demonstrates the Vision Advantage of Being Small

with Paul Vicinanza and Sameer B. Srivastava

Which groups are most likely to become visionaries that define the future of their field? Because vision is difficult to measure, prior work has reached conflicting conclusions: one perspective emphasizes the benefits of being large, established, and central, while another stresses the value of being small, upstart, and peripheral. We propose that this tension can be resolved by disentangling vision---the capacity to generate contextually novel ideas that foretell the future of a field---from the traces of vision that result in tangible innovation. Using Bidirectional Encoder Representations from Transformers (BERT), we develop a novel method to identify the visionaries in a field from conversational text data. Applying this method to a corpus of over 100,000 quarterly earnings calls conducted by 6,000 firms from 2011 to 2016, we develop a measure---prescience---that identifies novel ideas which later become commonplace. Prescience is predictive of firms' stock market returns: A one standard deviation increase in prescience is associated with a 4% increase in annual returns, and firms exhibiting especially high levels of prescience (above the 95th percentile) reap especially high returns. Moreover, contrary to theories of incumbent advantage, we find that small firms are more likely to possess prescience than large firms. The method we develop can be readily extended to other domains to identify visionary individuals and groups based on the language they use rather than the artifacts they produce.

Situated Cultural Fit: Value Congruence, Perceptual Accuracy, and the Interpersonal Transmission of Culture

with Richard Lu, Jennifer A. Chatman and Sameer B. Srivastava

Why are some people more successful than others at fitting in culturally over time? Prior research has offered divergent and seemingly inconsistent answers to this question. One perspective has highlighted the importance of shared values in shaping behavior, while another has emphasized the role of situational cues and the ability to read the group's cultural code. We develop a theoretical account that reconciles these competing perspectives. Drawing on dual-process theories of culture and cognition and the distinction between constrained and unconstrained situations, we develop a situated theory of cultural fit. We argue that values matter for behavior in unconstrained situations-in particular, for the choice to remain at or voluntarily exit from the organization. In contrast, perceptual accuracy matters for behavior in constrained situations-specifically, for the capacity to exhibit real-time linguistic conformity with peers. We further show that a person's behavior and perceptual accuracy are both influenced by observations of others' behavior, whereas value congruence is less susceptible to peer influence. Drawing on email and survey data from a mid-sized technology firm, we use the tools of computational linguistics and machine learning to develop longitudinal measures of cognitive and behavioral cultural fit. We also take advantage of a reorganization that produced quasi-exogenous shifts in employees' interlocutors to identify the causal impact of peer influence. We discuss implications of these findings for research on person-culture fit, cultural change and transmission, dual-process models of culture and cognition, and the pairing of surveys with digital trace data

Distinguishing Round from Square Pegs: Predicting Hiring Based on Pre-hire Language Use

with Sarah K. Stein and Sameer B. Srivastava

This article examines how cultural matching relates to a job applicant's likelihood of getting hired into an organization and identifies the components of cultural similarity that matter most for hiring success. Cultural compatibility at the hiring stage can forecast an individual's post-hire productivity but is diffi- cult to reliably measure in the selection process. As a consequence, cultural matching is often subject to various informational and identity-based biases. We develop a language-based model that provides a means for directly assessing job candidates' cultural similarity. Based on variegated data from a mid-sized technology firm-including job applicants' free text responses at the pre-hire stage, applicant characteristics, applicant-interviewer assignments, and hiring outcomes-we find that linguistic similarity with previously hired employees increases a job candidate's chances of being hired, even after controlling for the applicant's human and social capital. We further find that, although all three forms of cultural fit that we assess-fit based on work preferences, lifestyles, and ideology-predict hiring in between-interviewer models, only work preferences fit predicts hiring in within-interviewer models. Supplemental analyses indicate that pre-hire cultural fit is also predictive of successful enculturation in the firm over the first six months of employment. Together, these results indicate that cultural matching leads to sorting on attributes that are both relevant and potentially irrelevant for job success.

Network Boundedness as Market Identity: Evidence from the Film Industry

with Anthony Vashevko

Network scholars have studied organizational creativity predominantly as a problem of acquiring and integrating information. In contrast, we re-conceptualize the structural tradeoff between brokerage and closure as an identity signal. With Hollywood as our empirical setting, we demonstrate that consumer perceptions of films are structured by a strong status hierarchy, and that boundedness - the extent to which a film's production team comprises an exclusive clique within the network of interpersonal collaborations in the film industry - serves as a signal of artistic quality. We draw on a uniquely detailed dataset of consumer preferences, and take advantage of the lag between film production and consumer evaluation as a means to demonstrate that production team members' career trajectories after a film had been produced have a bearing on audiences' evaluations. Films whose team members went on to collaborate in exclusive circles and thereby, we argue, establishing a high-status identity, tend to enjoy greater post-hoc artistic appreciation than at the time of their release.