Critical Analysis of Management & Marketing, Volumes 18–19 (2023–2024)

Dimension 1: Methodological Rigor

Convenience Sampling Epidemic

A disturbing proportion of the empirical papers in this corpus rely on convenience or non-probabilistic sampling—and openly state as much in their abstracts. The following papers explicitly declare convenience, snowball, non-probabilistic, or purposive sampling:

That is at minimum 6 papers with explicitly non-probabilistic sampling. The real number is higher: many papers that describe “online surveys” distributed through social media or personal networks are de facto convenience samples without acknowledging it. Malik et al. (2023, 0026) collected data “using Google Forms from individuals who have used social commerce sites”—a self-selected online sample with no probabilistic design. Nicolae (2024, 0016) surveyed “595 urban consumers through an online survey” without describing any sampling frame. Pallikkara, Pinto & Hawaldar (2024, 0012) describe a “cross-sectional survey involving 385 participants” with no mention of sampling method. The pattern is clear: convenience sampling is the default, not the exception.

PLS-SEM as Methodological Monoculture

Partial Least Squares Structural Equation Modeling (PLS-SEM), overwhelmingly executed through SmartPLS software, dominates the empirical toolkit of this corpus. Papers using PLS-SEM or SmartPLS:

That is 12 papers using PLS-SEM—a single technique accounting for a massive share of the empirical work. When additional SEM papers are counted (those using CB-SEM or AMOS), the structural equation modeling monoculture expands further: Pham & Vu (2024, 0007) used covariance-based SEM, Pallikkara et al. (2024, 0012) used SEM, Matíc et al. (2024, 0019) used SmartPLS with AMOS, and García-Valenzuela, Jacobo-Hernandez & Flores-López (2023, 0027) used CB-SEM. This means roughly 16 papers—a clear majority of the survey-based empirical work—deploy some variant of structural equation modeling. The methodological vocabulary of this journal is alarmingly narrow.

Small Samples vs. Grand Claims

Several papers draw sweeping conclusions from sample sizes that are small relative to their variable count or generalization scope:

Zero Experimental Designs

Not a single paper in the entire 64-paper corpus employs a true experimental design with random assignment to conditions. The entire body of empirical work is observational: surveys, archival data analysis, bibliometric reviews, and case studies. This means no paper can legitimately make causal claims—yet many do (see below).

Self-Reported Data Dominance

The overwhelming majority of empirical papers rely exclusively on self-reported questionnaire data. The exceptions using non-survey data are few and notable:

Everything else—the vast majority—relies on people reporting their own attitudes, intentions, perceptions, and behaviors through questionnaires. Common-method bias is a systemic threat that almost none of these papers address in their abstracts.

Cross-Sectional Designs with Causal Language

Papers using cross-sectional survey designs routinely employ causal language (“influence,” “impact,” “affect,” “effect”) that their research design cannot support. Examples are pervasive:

This is not a minor terminology issue. These papers uniformly employ directional arrows in SEM path diagrams and report “effects” from cross-sectional snapshots. Correlation is presented as causation throughout. The editorial process has failed to enforce the most basic distinction in empirical research.

Dimension 2: Statistical Concerns

Beta Values Without Effect Size Context

Cuong (2024, 0002) reports individual beta values to three decimal places in the abstract itself (β = 0.502, β = 0.217, β = 0.207, β = 0.758, β = 0.822), presenting them as evidence of “favorable impact” and “strong positive influence”—with zero discussion of effect sizes, explained variance, or practical significance. Is a β of 0.207 practically meaningful for online retailers? The paper does not say. This pattern of reporting standardized coefficients without R² values, Cohen’s f², or confidence intervals characterizes the SEM-heavy papers throughout the corpus.

Confirmatory-Only Approaches

The corpus is dominated by confirmatory hypothesis testing. Papers formulate hypotheses from existing literature, collect a single dataset, test the hypotheses, and report whether they are “supported” or “not supported.” Not one paper in the corpus reports an exploratory-confirmatory split-sample design. Not one includes a replication sample. Not one reports a pre-registered hypothesis. The entire corpus is a post-hoc confirmatory exercise, and the near-universal result is hypothesis confirmation—a pattern consistent with publication bias or analytical flexibility rather than genuine theoretical precision.

SEM Model Fit: Unreported or Questionable

The PLS-SEM papers face a particular issue: PLS-SEM does not produce the global fit indices (CFI, RMSEA, SRMR) that CB-SEM provides, making model evaluation less rigorous by design. The abstracts of the PLS-SEM papers (Koe et al. (0003), Purnama et al. (0010), Guo et al. (0011), Khalid (0017), Keller et al. (0022), Alves et al. (0031)) do not mention any model evaluation criteria. The reader is asked to trust that the model is adequate based on the word “findings.”

Multiple Testing Without Correction

Papers testing numerous path relationships simultaneously—Cuong (2024, 0002) tests multiple mediational paths, Sembel et al. (2024, 0025) tests seven independent variables on adoption behavior—show no evidence of correcting for multiple comparisons. Given that SEM decomposes models into many individual path tests, the inflation of family-wise Type I error is a concern that the corpus ignores entirely.

Statistical Significance as the Only Metric

Úsas, Jasinskas & Streimikiene (2024, 0032) conclude that “website quality, and satisfaction positively affect customer loyalty” based on their SPSS regression analysis of 400 respondents, without any mention of effect magnitudes. Karamanis et al. (2024, 0005) report “a statistically significant positive correlation” between GDP and healthcare expenditure—a relationship so obvious it requires effect size context to have any analytical value. The p-value is treated as a badge of scientific credibility rather than one component of a comprehensive statistical assessment.

Dimension 3: Novelty and Contribution

“First Study to...” Claims

Multiple papers assert uniqueness claims that strain credulity:

Bibliometric and Review Papers: Synthesis Without Generation

The corpus contains a notable proportion of papers that synthesize existing literature rather than produce new empirical findings:

That is 6 out of 64 papers (nearly 10%) that produce no original data. Bibliometric analysis has become a low-effort publication strategy—run VOSviewer on a Scopus query, describe the clusters, propose a “future research agenda.” These papers occupy journal space without advancing the field’s empirical frontier.

TAM/TPB/UTAUT Country-Context Replications

A significant cluster of papers applies well-established theoretical frameworks to new country contexts without extending or challenging the underlying theory:

The formula is standardized: take an established framework, add a country name, collect convenience sample surveys, run PLS-SEM, confirm the hypotheses. The knowledge increment from any individual paper is negligible because these frameworks have been tested in dozens of countries already. The only “novelty” is the geographic label.

Incremental vs. Genuine Advances

The papers that stand apart methodologically are a small minority: Zuo et al. (2024, 0008) with its multimodal AI-driven sentiment analysis of emojis; Khowaja et al. (2024, 0026) with its Local Parametric Approach to M&A forecasting; Seo & Inoue (2023, 0014) with 11,863 records of actual video game sales; and Campeanu, Boitan & Anghel (2023, 0017) using LASSO regression and machine learning. These represent genuine methodological diversity. The remaining 60 papers overwhelmingly follow the survey-SEM-confirm template.

Dimension 4: Geographic & Institutional Bias

Country Concentration

The corpus exhibits heavy geographic clustering. Based on author affiliations, study populations, and explicit country references in abstracts, the dominant countries are:

Romania and Hungary together account for approximately 28 of 64 papers—nearly half the corpus. This is a journal headquartered in Romania publishing predominantly Romanian and Hungarian research. The international reach is thin: Southeast Asian papers come from Vietnam, Indonesia, Malaysia, and Thailand; Western European papers from Portugal, Spain, and Greece; one from Colombia; one from Japan. North America, the UK, Scandinavia, and Africa are entirely absent as study contexts (though Bratianu & Paiuc (2023, 0006) compares USA and Romania).

Country-Specific Findings Presented as Universal

Cuong (2024, 0002) studies 294 Vietnamese e-commerce consumers and concludes with advice for “online retailers” generally. Koe et al. (2024, 0003) studies Malaysian MSMEs and claims the institutional model “should not be neglected in fostering SE intention”—unqualified by geography. Khalid (2024, 0017) surveys 442 Thai respondents and offers implications for “marketing managers” with no localization caveat. This framing of country-specific convenience samples as universally applicable is a recurring failure of academic discipline.

Recurring Author Teams

The team of Úsas, Jasinskas & Streimikiene publishes in both Volume 18 (2023, DOI 0007) and Volume 19 (2024, DOI 0032) on the same topic (C2C e-commerce in Lithuania). Juhász appears as a co-author in both Bencsik et al. (2024, 0004) and Molnár et al. (2024, 0015). Stamule appears in both Păunescu et al. (2023, 0012) and Birgen et al. (2024, 0027). Davidescu and Manta appear together in both Davidescu et al. (2024, 0014) and Manta et al. (2023, 0019). This raises questions about whether the journal serves a rotating network of repeat contributors rather than a diverse scholarly community.

Language Quality Signals

Several abstracts contain grammatical constructions suggesting inadequate copy-editing: Úsas et al. (2024, 0032) writes “Descriptive statistics and regression analyses was used” (subject-verb disagreement). Úsas et al. (2023, 0007) writes “what the user wants and what problems he faces” (gendered pronoun). Prada & Cimpoeru (2023, 0032) has “Using data from the Eurostat database, we undertake an extensive analysis of migration movements” followed immediately by identical phrasing. These are not isolated lapses; they appear in final published versions, suggesting the editorial and production pipeline lacks adequate English-language quality control.

Dimension 5: Journal-Level Critique

Topical Incoherence: Management Journal, Marketing Journal, or Catch-All?

A journal titled Management & Marketing should publish research on management and marketing. A substantial number of papers in this corpus have no discernible connection to either discipline:

That is 16 of 64 papers—a full 25%—whose connection to management or marketing is tenuous at best, non-existent at worst. Hybrid warfare education, biomass value chains, heat pump adoption, NUTS2-level migration spatial econometrics, and capital market volatility are topics that belong in specialized energy, migration, or finance journals. Their presence here indicates a journal that will publish virtually anything submitted by authors within its institutional network, regardless of topical fit.

Scope Discipline

The evidence above answers the question definitively: editorial gatekeeping on topical scope is either absent or ineffective. A journal that publishes papers on hybrid warfare training and black pellet coal substitution alongside papers on guerrilla marketing and brand equity has no coherent identity. This is a generalist social science outlet masquerading as a specialized journal.

Special Issue Coherence (2023)

The 2023 Special Issue (Vol. 18, SI) contains the following papers:

This “Special Issue” covers student engagement, metaverse workplace design, direct taxation bibliometrics, global energy mix forecasting, urban development efficiency, renewable energy, cashierless retail, hospitality wages, Vietnamese SME digital transformation, and social commerce. There is no coherent theme. This is a miscellany labeled as a Special Issue—a practice that undermines whatever thematic signal a Special Issue designation is supposed to convey.

Peer Review Signals

The quality variation across papers is extreme. The same journal publishes Khowaja et al. (2024, 0026)—a technically sophisticated paper on count-data time series with Local Parametric Approach and Multiplier Bootstrap—alongside Úsas et al. (2024, 0032)—a paper whose abstract contains subject-verb disagreement and whose entire contribution is that “website quality, and satisfaction positively affect customer loyalty” based on SPSS regression of 400 respondents. When a journal’s output ranges from genuinely novel methodological contributions to basic SPSS regressions confirming the obvious, the peer review process is not functioning as a quality filter. It is functioning as a throughput pipeline.

Dimension 6: Specific Red Flags

Dimension 7: Writing and Presentation Quality

Abstract Length Variation

Abstract lengths vary wildly, from approximately 80 words (Sánchez-Cubo & Mondéjar-Jiménez, 2023, 0024—a truncated abstract that barely describes the method) to over 350 words (Jora et al., 2023, 0002 on space economy; Nicolae, 2024, 0016 on sustainable purchasing). There is no enforced word limit. Some abstracts read as executive summaries; others read as introductory paragraphs. A journal without abstract formatting standards signals a production process without templates or editorial enforcement.

Keyword Quality Issues

Keywords across the corpus display multiple quality problems:

Title Informativeness

Title quality spans a wide spectrum:

The absence of title formatting guidelines is evident. No editorial policy enforces conciseness, informativeness, or stylistic consistency.

Summary Assessment

This 64-paper corpus reveals a journal with systemic structural weaknesses across every evaluative dimension. The methodological base is narrow (PLS-SEM monoculture, convenience sampling, zero experimental designs), the statistical reporting is superficial (beta coefficients without effect sizes, causal language from cross-sectional data), and the novelty threshold is low (country-context replications of established frameworks, bibliometric reviews as a publication strategy). Geographic concentration in Romania and Hungary, combined with recurring author teams and topical incoherence (hybrid warfare, biomass, heat pumps, space economy), indicates a journal functioning as a regional publication outlet rather than a competitive international forum. The production quality failures—wrong titles, corrupted citations, inconsistent DOIs, duplicate citation lines—are not cosmetic: they corrupt the bibliographic record and undermine the journal’s professional credibility. The evidence of salami-slicing by the Úsas/Jasinskas/Streimikiene team, undetected across two volumes, is a peer review failure of the first order.

For a tenure and promotion committee evaluating publications in this journal: the venue-level quality signals warrant careful scrutiny of any individual paper before assigning it weight in an academic dossier.

Evidence Audit: Distinct Papers Cited

This analysis cites the following distinct papers as evidence (by DOI suffix and primary author), exceeding the minimum threshold of 20:

Total distinct papers cited: 62 out of 64. The two papers not individually named as evidence in a critique point are López-Rodríguez et al. (2024, 0006) and Lepik & Sakarias (2023, 0003), both of which do not exhibit the specific methodological or production issues catalogued above but belong to the same corpus context.