Participatory practices at work change attitudes and behavior toward societal authority and justice

The study asks whether generalized attitudes toward authority and justice—often treated as stable, personality-like dispositions—can change over the short term through local participatory experiences at work. Drawing on theories of participatory democracy (Pateman) and Lewin’s notion of groups as “cultural islands,” the authors hypothesize that brief, immersive participatory practices within routine workplace meetings will decrease deference to generalized authority, reduce belief in a just world, and increase willingness to participate in decision-making beyond work. The research tests these ideas in two very different contexts (a Chinese factory and a U.S. university) to assess the causal impact and generality of workplace participation on broader democratic attitudes and empowerment.

Classic work conceptualizes authoritarianism as a durable personality trait shaped early in life and reinforced by long-term socialization (e.g., Adorno, Altemeyer), with structural correlates such as societal threats linked to higher authoritarian attitudes. These generalized attitudes are typically seen as stable over the lifespan and distinct from attitudes toward specific authorities, which are more malleable. Political and economic theorists (e.g., Smith, Marx) argued that long-term work organization shapes generalized attitudes toward authority and justice, suggesting lower socioeconomic groups may be socialized into subservience. In contrast, Pateman’s participatory democracy theory posits that participatory workplace structures can cultivate political efficacy and less deference to authority and hierarchy. Lewin theorized that small groups can form “cultural islands,” where immersive democratic procedures foster new norms and behavior. Empirical tests have largely been observational, correlating perceived workplace participation with democratic attitudes, with mixed results and a focus on Western settings. There is little to no experimental evidence, especially in non-democratic contexts, on whether local participatory practices causally change generalized attitudes toward societal authority and justice. This gap motivates the present field experiments.

Design: Two preregistered field experiments manipulated weekly participatory meetings in existing work groups for six weeks and measured outcomes weeks after the intervention. Study 1 took place in a multinational textile factory in China; Study 2 in administrative staff groups at a U.S. university.

Participants and randomization:

• Study 1 (China): 65 sewing work groups (N=1,752 workers; 93.6% female; mean age 32.5). Groups were randomized via non-bipartite optimal matching to treatment (participatory meetings) or control (observer of status quo meeting). Building layout and work routines minimized intergroup spillover. Informed consent obtained.
• Study 2 (U.S.): 32 academic departments’ administrative groups (N=172 staff; 78% female; mean age 50; 80% White/European-American). Groups randomized to treatment (participatory meetings) or control (status quo meetings). Informed consent obtained.

Intervention (6 weekly sessions, 20 minutes each):

• Study 1 treatment: A trained research assistant (RA) facilitated a structured participatory discussion in the presence of the supervisor, who was instructed not to speak or interrupt. Prompts focused on work challenges, strategies, and goals; all workers were encouraged to contribute. Meetings ended with workers publicly stating individual production goals.
• Study 1 control: An RA silently observed the existing supervisor-led meeting (lecture on performance and strategies; goals set by supervisor). No changes were encouraged.
• Study 2 treatment: For logistical reasons, supervisors themselves led participatory meetings following a protocol closely translated from Study 1, with emphasis on active listening and non-interruption, discussion of work challenges/strategies, and goal setting.
• Study 2 control: Status quo meetings continued; no observer present.

Measurement timing and instruments:

• Study 1: Four weeks post-intervention, individual surveys (completion rate 84.07%). Outcomes: generalized attitudes toward authority (e.g., obedience/respect items), belief in a just world (e.g., people deserve what they get), perceived intergroup conflict (rich vs ordinary; capitalists vs working class; managers vs workers), and participation outside work (political interest; family/social decision-making). Six-point Likert scale; items adapted from established scales; translations/back-translations; EFA/CFA supported indices. Demographics (age, gender, marital status, rural/urban origin) collected one week after intervention.
• Study 2: Two weeks post-intervention, surveys measured generalized attitudes toward authority and belief in a just world on a 7-point Likert scale. Perceived conflict and participation outside work were not measured due to time constraints.

Analysis:

• Primary analyses estimated average treatment effects via linear regressions with department fixed effects, a treatment indicator, and pre-treatment demographics as covariates; robust standard errors clustered by group. Joint significance tests assessed whether sets of treatment effects differed from zero across outcomes. Robustness checks included group-mean t-tests (N=65 groups in Study 1), item-level analyses, and post-hoc power analyses (Study 1 power=0.99; Study 2 power=0.82).

Across 97 groups and 1,924 participants, brief participatory meetings causally shifted generalized attitudes toward authority and justice and related behaviors.

Study 1 (China, N=1,752):

• Generalized attitudes toward authority: Treatment groups reported lower deference to authority (M=3.87, SD=0.32) vs control (M=4.23, SD=0.33); β=-0.39, 95% CI [-0.55, -0.23], SE=0.08, p<0.001.
• Belief in a just world: Lower in treatment (M=3.86, SD=0.22) vs control (M=4.10, SD=0.16); β=-0.26, 95% CI [-0.34, -0.18], SE=0.04, p<0.001.
• Perceived intergroup conflict: No significant differences for rich vs ordinary (p=0.24) or capitalists vs working class (p=0.10). Higher perceived conflict between managers and workers in treatment (M=3.55, SD=0.27) vs control (M=3.29, SD=0.17); β=0.31, 95% CI [0.21, 0.42], SE=0.05, p<0.001.
• Participation outside of work: Higher overall participation in treatment (M=4.39, SD=0.19) vs control (M=4.21, SD=0.19); β=0.18, SE=0.05, 95% CI [0.03, 0.21], p<0.001. Greater political interest (M=4.06 vs 3.80); β=0.29, 95% CI [0.13, 0.45], SE=0.08, p<0.001. Greater family/social participation (M=4.54 vs 4.41); β=0.12, 95% CI [0.03, 0.21], SE=0.05, p=0.012.
• Joint significance across outcomes: F1,58=8.06, p<0.001.

Study 2 (U.S., N=172):

• Generalized attitudes toward authority: Lower in treatment (M=2.70, SD=0.88) vs control (M=2.90, SD=0.46); β=-0.44, 95% CI [-0.85, -0.03], SE=0.21, p=0.037.
• Belief in a just world: Lower in treatment (M=3.93, SD=0.41) vs control (M=4.18, SD=0.35); β=-0.23, 95% CI [-0.45, -0.005], SE=0.11, p=0.045.
• Joint significance across attitudinal outcomes: F1,26=3.35, p=0.039.

Effect sizes were similar across settings despite contextual differences. Changes persisted two to four weeks after the intervention ended.

Findings demonstrate that brief, structured participatory experiences within existing workplace groups can causally and durably reduce generalized deference to authority and belief in a just world—attitudes often regarded as stable products of long-term socialization. The effects appeared in a highly hierarchical, non-democratic context (Chinese factory) and in a democratic, higher-education context (U.S. university), suggesting generality across cultures and work environments. In Study 1, participants simultaneously increased positive views of their local factory management (reported elsewhere) and became more critical of societal authority, indicating differential effects on local vs generalized authority perceptions consistent with procedural justice for local authorities and broader empowerment beyond the immediate context. Increased self-reported political interest and family/social participation among Chinese workers further supports a pattern of empowerment and spillover into civic and interpersonal domains. These results provide rare causal support for participatory democracy theory and extend Lewin’s “cultural islands” idea by showing spillover from local participatory practices to generalized attitudes, not just local behavior. The work underscores that modest, regular opportunities for voice—in this case, 20-minute weekly meetings—can shift social attitudes thought to be personality-like, with implications for fostering democratic orientations in varied organizational settings.

The study offers causal evidence that introducing brief participatory practices into routine workplace meetings can reduce authoritarian attitudes and belief in a just world and increase participation tendencies beyond work. These effects persisted weeks after the intervention and generalized across a non-democratic industrial setting and a democratic university context. The findings advance participatory democracy theory and suggest that everyday, immersive opportunities for voice may cultivate broader democratic attitudes and empowerment without overhauling institutional structures. Future research should measure concrete behavioral outcomes beyond self-report, test the necessity of workers’ lower-status positions for effects to emerge, examine longer-term persistence, explore school and other institutional settings, and investigate how local participatory norms diffuse to civic and political engagement.

• Outcomes largely rely on self-reported attitudes and behaviors; concrete behavioral measures (e.g., observed civic actions, behavioral tasks) were not collected.
• Follow-up periods were short (2–4 weeks); long-term durability is unknown.
• Study 2 did not assess perceived intergroup conflict or participation outside of work due to survey time constraints, limiting cross-study comparability.
• Protocol differences across settings (RAs facilitated in Study 1; supervisors facilitated in Study 2) may affect mechanisms and generalizability.
• Samples were drawn from lower-status support roles; effects may differ for higher-status employees or other sectors.
• Cultural and scale differences (6-point without neutral in China vs 7-point in U.S.) may influence response distributions, though analyses used within-study comparisons.
• While building layout and routines minimized spillovers in Study 1, unmeasured intergroup communication cannot be entirely ruled out.