APA Business Literature Review Example: Diversity Training Effectiveness

5 business–sociology scoring dimensions (combined with APA)

Dimension Poor signal Strong signal

1 Outcome distinction "Training works / doesn't work" Separates reactions, cognitive learning, attitudinal/affective learning, and behavior

2 Effect-size reporting No magnitude reported Names the meta-analytic effect (e.g., Hedges' g = .38) and its confidence interval

3 Design hierarchy Anecdotes and editorials only Meta-analysis (Level I) > field experiment > quasi-experiment > self-report survey

4 APA citation mix All narrative or all parenthetical ≥40% narrative citations; et al. italicized and introduced at first citation

5 Precise gap "More research is needed" Names population × context × outcome × time horizon

Strong draft (APA 7th, annotated)

[Thesis / scope] Workplace diversity training produces reliable short-term gains in reactions to and cognitive learning about diversity, but its effects on attitudes and behaviors are smaller, decay over time, and depend on how the training is embedded in organizational practice. [Narrative, 4 authors + meta-analysis] Bezrukova et al. (2016) synthesized 260 independent samples and reported an overall effect size of Hedges' g = .38—moderate in magnitude—with the largest effects for trainees' reactions and cognitive learning and markedly smaller effects for attitudinal/affective and behavioral learning. [Parenthetical, multi-source, alphabetical] Kulik and Roberson (2008) anticipated this pattern in their review of diversity education evaluations, while Leslie (2019) situated it within a broader review of diversity, discrimination, and inclusion at work. [Counter-evidence / field experiment] The attitude–behavior gap is clearest in Chang et al.'s (2019) preregistered field experiment (N = 3,016) at a global organization: although the brief online training changed attitudes among groups with initially less favorable views toward women, it did not change their workplace behavior; conversely, limited behavior change emerged only among those already strongly supportive before training. [Structural critique] At the organizational level, Dobbin and Kalev (2016) argued that standalone training can backfire when it is mandatory and unaccompanied by structural reform, a critique Noon (2018) extended by questioning the agency assumptions underlying unconscious-bias interventions. [Mechanism / synthesis] Taken together, the evidence supports a conditional conclusion: diversity training is most effective when it targets both awareness and skill development, is delivered over a substantial duration, and is complemented by other diversity initiatives—the moderators Bezrukova et al. (2016) identified. [Precise gap] What remains untested is whether these effects generalize to small and medium-sized enterprises (SMEs) in non-Anglophone contexts, measured via longitudinal behavioral outcomes (e.g., promotion and retention) rather than self-reported attitudes—a gap that matters because the existing meta-analytic and field-experimental base is concentrated in large, English-speaking organizations.

[Evidence-quality subsection] Across the literature, design quality determines confidence. [Narrative, meta-analysis] Bezrukova et al.'s (2016) Level I meta-analysis is the evidentiary anchor: its 260 samples span four outcome classes and reveal that reactions and attitudinal/affective learning decay over time, whereas cognitive learning remains stable and in some cases increases (Bezrukova et al., 2016). [Parenthetical, field experiment] Chang et al. (2019) supplied the strongest causal evidence—a preregistered experiment with an active placebo control and 20-week follow-up—but its outcome was still constrained to self-report and a limited set of real workplace decisions. [Synthesis] A review-level reading therefore supports moderate certainty for cognitive and attitudinal outcomes but only low certainty for durable behavioral change and objective workforce metrics, where field studies remain sparse. [Gap restatement] Future reviews should separate same-source self-report designs from multi-source and behavioral designs before pooling effects.

[Implementation subsection] Translation is the second gap. [Narrative, 1 author] Noon (2018) cautioned that unconscious-bias training overstates individual agency and underweights structural constraints. [Parenthetical, 2 authors] Consistent with this, Dobbin and Kalev (2016) documented that voluntary, skill-based, long-duration programs embedded in broader initiatives outperform one-off compliance workshops. [Precise gap] A literature review that treats training as a universal intervention without modeling mandatory status, duration, and complementary practices will systematically overstate transferability to organizations lacking those conditions.

In-text citation audit (APA 7th demonstrated)

• 1 author: Leslie (2019) / (Leslie, 2019)

• 2 authors: Kulik and Roberson (2008) / (Kulik & Roberson, 2008) — & in parenthetical, "and" in narrative

• 3+ authors, first citation: Bezrukova, Spell, Perry, and Jehn (2016) / (Bezrukova et al., 2016) — full list first time only

• 3+ authors, subsequent: (Bezrukova et al., 2016) — italicized et al.

• 6+ authors: Chang et al. (2019) — first citation already uses et al.

• Multi-source parenthetical: (Chang et al., 2019; Kulik & Roberson, 2008; Leslie, 2019) — alphabetical by first author, semicolon-separated

• Direct quote with page: (Bezrukova et al., 2016, p. 1235) — comma before p.

• Narrative with own emphasis: Bezrukova et al. (2016) noted that...

Reference list (APA 7th, matches citations above)

Bezrukova, K., Spell, C. S., Perry, J. L., & Jehn, K. A. (2016). A meta-analytical integration of over 40 years of research on diversity training evaluation. Psychological Bulletin, 142(11), 1227–1274. https://doi.org/10.1037/bul0000067

Chang, E. H., Milkman, K. L., Gromet, D. M., Rebele, R. W., Massey, C., Duckworth, A. L., & Grant, A. M. (2019). The mixed effects of online diversity training. Proceedings of the National Academy of Sciences of the United States of America, 116(16), 7778–7783. https://doi.org/10.1073/pnas.1816076116

Dobbin, F., & Kalev, A. (2016). Why diversity programs fail. Harvard Business Review, 94(7/8), 52–60.

Kulik, C. T., & Roberson, L. (2008). Common goals and golden opportunities: Evaluations of diversity training programs. Academy of Management Learning & Education, 7(3), 309–331. https://doi.org/10.5465/amle.2008.34251664

Leslie, L. M. (2019). Diversity, discrimination, and workplace inclusion: A review and integration. Annual Review of Organizational Psychology and Organizational Behavior, 6, 227–252. https://doi.org/10.1146/annurev-orgpsych-012218-024757

Noon, M. (2018). Pointless diversity training: Unconscious bias, new racism and agency. Work, Employment and Society, 32(1), 198–209. https://doi.org/10.1177/0950017017719841

Format checks (APA 7th): • Heading "References", centered (not "Bibliography" or "Works Cited")

• Hanging indent; authors last name, initials; ampersand (&) before final author

• Sentence case for article titles; italics for journal names (title case) and volume numbers

• Issue number in parentheses with no space: 142(11); comma before page range

• DOI as URL: https://doi.org/... — no label, no period at the end

• Alphabetical by surname: B → C → D → K → L → N

• et al. italicized throughout

What changed: poor draft → strong APA draft

Element Poor draft Strong APA draft (above)

Opening "Diversity training is good for companies." Scope + four outcome classes + conditional thesis

Effect size "Studies show mixed results." Hedges' g = .38, largest for reactions/cognitive learning

Evidence hierarchy Primary studies only Meta-analysis (Level I) + preregistered field experiment separated

Attitude–behavior gap Omitted Central mechanism: Chang et al. (2019) attitude change ≠ behavior change

Citation mechanics Harvard-style (Author Year) (Author, Year) + italicized et al. + & in parentheses

Multi-source One source per sentence Alphabetical within parentheses, semicolon-separated

Counter-evidence Omitted Dobbin & Kalev (2016) + Noon (2018) structural critique

Gap "More research needed." SME × non-Anglophone × longitudinal behavioral outcomes

Reference list Harvard "Reference list" APA "References" with DOIs as URLs

5 fatal signals (business/sociology-specific)

1. The review reports only attitudinal outcomes as if they proved behavioral change—Chang et al. (2019) showed they can diverge.

2. Every claim is (Author, Year) at sentence end—zero narrative citations.

3. et al. is not italicized (Harvard/MLA habit) or appears only on second citation of a 3+ author work.

4. The list is titled "Works Cited" or "Bibliography" instead of "References."

5. The review treats training as a black box—mandatory vs. voluntary, duration, awareness vs. skill focus, and complementary initiatives are never modeled (Bezrukova et al., 2016, found all four to be moderators).

Rewrite Checklist (APA × Business-Soc)

1. Open by naming the four outcome classes (reactions, cognitive learning, attitudinal/affective learning, behavior)—not "does it work?"

2. Anchor the quantitative claim with Bezrukova et al. (2016): Hedges' g = .38 overall; cognitive learning largest; behavioral smallest and time-decaying.

3. State the attitude–behavior gap explicitly, using Chang et al. (2019)'s preregistered N = 3,016 field experiment.

4. Convert ≥40% of parenthetical citations to narrative (author as subject).

5. Audit every 3+ author work: italicized et al., introduced at first citation with all names.

6. Include the structural critique (Dobbin & Kalev, 2016; Noon, 2018): training alone, especially mandatory one-offs, may backfire.

7. Close with a gap naming population × context × outcome × time horizon (e.g., SMEs, non-Anglophone, 12-month retention/promotion data).

8. Build the References list in parallel; verify every DOI resolves; apply APA 7th hanging indent and DOI-as-URL formatting.

FAQ

Why report Hedges' g and not just "the effect was significant"? Because the size matters for practice. Bezrukova et al. (2016) reported Hedges' g = .38, a moderate effect—but it was not uniform: cognitive outcomes were strongest, behavioral outcomes weakest. Reporting the metric lets your reader judge whether the intervention is worth the cost, rather than just "significant vs. not."

What is the attitude–behavior gap? Chang et al. (2019) found that a brief online training changed attitudes among participants with initially less favorable views toward women, but not their behavior; limited behavior change appeared only in those already strongly supportive. This is why a review must never infer behavioral change from attitude surveys alone.

Is APA the right format for a business literature review? In the US, APA is common for management, organizational behavior, and HRM programs, and it is the default across most social sciences. In the UK, Australia, and much of Europe, Harvard often replaces APA for business (see /harvard-business-literature-review-example). If you want to compare the same topic in both formats, those two pages are designed to be read together.

How do I cite a meta-analysis vs. a field experiment? Both use standard APA author–year citations. The distinction belongs in your narrative, where you grade evidence: meta-analyses (like Bezrukova et al., 2016) provide population-level estimates; preregistered field experiments (like Chang et al., 2019) provide stronger causal claims for a specific intervention. Name the design explicitly so the reader can weight it.

Can I reuse this review's citations? No. Use the structure and APA mechanics as a model; replace the sources with your own verified search results, and confirm each DOI resolves.

/apa-7th-literature-review-example · /literature-review-example-annotated-sample#by-citation-style · /harvard-psychology-literature-review-example · /literature-review-example-annotated-sample

Continue the research workflow with Acade

Acade can help you search for literature, organize evidence notes, build a synthesis outline, and revise citation-supported academic writing. It cannot guarantee source accuracy or replace your reading, methodological judgment, institutional rules, or responsibility for the submitted work.

Tell Acade your topic, discipline, citation style, evidence requirements, and assignment constraints, then ask it to help organize the next verifiable research step.