The research team, Erkut Altindağ of Doğuş University and Özge Gül of Istanbul Rumeli University, collected 42,852 public LinkedIn comments from January to December 2023. They targeted discussions on posts tagged with keywords such as compensation, benefits, salary, and total rewards, along with conversations in professional human resources groups and responses to corporate announcements about benefits packages. Rather than relying on the structured questionnaires and interviews that have long dominated compensation research—methods vulnerable to social desirability bias and flawed retrospective recall—the researchers tapped into unfiltered, spontaneous discourse. They are careful, however, to frame social media as a complement to surveys rather than a replacement, acknowledging that LinkedIn participation is shaped by self-selection, vocal minorities, and the performative nature of public posting.

Technically, the study deployed a multi-modal analytical pipeline built on some of the most powerful tools in modern computational linguistics. At its core was a BERT-based sentiment analysis model, fine-tuned on 1,000 manually labeled compensation-related comments to classify each remark as positive, neutral, or negative. The BERT architecture, a deep bidirectional transformer encoder, outperformed simpler baselines including a TF-IDF logistic regression model (F1 = 0.71) and a lexicon-based VADER approach (F1 = 0.65), achieving a validation accuracy of 0.89 and an F1-score of 0.87 on the test set. Class imbalance in the training data was handled through class-weighted cross-entropy loss, and the framework incorporated a custom lexicon of compensation-specific terms to sharpen accuracy in this particular domain.

Alongside sentiment analysis, the researchers applied two complementary techniques for uncovering structure in the discourse. Topic modeling was performed using both Latent Dirichlet Allocation, optimized to 15 topics through coherence scoring, and BERTopic, a neural approach capable of tracking how themes evolved dynamically across the year. In parallel, a custom named entity recognition model built on spaCy’s framework and trained on 2,500 annotated comments extracted specific entities from the text: organizations, benefit types, and job roles. The entity recognition model achieved an F1-score of 0.83, with particularly strong performance in identifying organizations (F1 = 0.86) and benefit types (F1 = 0.83). Data quality was protected through rigorous preprocessing—English-language filtering using langdetect, deduplication via exact and fuzzy matching based on Levenshtein distance, and thorough text cleaning—while validation included inter-rater reliability assessment by three independent coders, who achieved a Cohen’s Kappa of 0.82, and five-fold cross-validation across all models.

The results paint a vivid picture of sector-specific sentiment in the world of work. Technology sector discussions maintained the highest positive sentiment, with a mean score of 0.42, followed by healthcare at 0.31 and financial services at 0.25. A one-way analysis of variance confirmed that these differences between industries were statistically significant (F(2, 42,849) = 23.45, p < 0.001). Sentiment also fluctuated throughout the year, with notable peaks coinciding with major corporate benefits announcements, and the technology sector showed greater volatility than other industries. The authors note that because LinkedIn comments are nested within users, posts, and organizations, the standard ANOVA violates independence assumptions, so the reported statistic should be treated as a first-order approximation, with multilevel modeling recommended for future work.

Perhaps the most striking finding concerns how the themes of employee conversation shifted over the course of a single year. Work-life balance discussions surged from 14.2 percent of the discourse in the first quarter of 2023 to 41.2 percent by the fourth quarter—a 27 percentage-point increase in thematic prevalence. Meanwhile, traditional base compensation discussions declined by 12.0 percent. A chi-square test confirmed the statistical significance of the overall thematic shift (χ²(4) = 156.23, p < 0.001), though the researchers emphasize that the omnibus test does not establish the significance of any single theme’s change; the individual percentage shifts are presented as descriptive magnitudes pending confirmatory pairwise comparisons with multiple-testing correction. Even with those caveats, the direction is clear: employees are talking less about raw salary and more about time, flexibility, and well-being.

The named entity recognition analysis added another layer of insight by linking specific benefits to sentiment outcomes. The model identified 3,427 unique organizations, 892 granular benefit-related entities—which aggregate onto roughly 47 canonical benefit categories—and 1,245 job roles across the dataset. Health insurance emerged as the benefit type most strongly correlated with positive sentiment (r = 0.42, p < 0.001, 95 percent confidence interval 0.38 to 0.46), followed closely by flexible work arrangements (r = 0.38, p < 0.001). These correlations remained robust after controlling for industry sector and temporal variation. Health insurance dominated the discourse with 12,458 mentions, while retirement benefits, despite lower frequency at 7,892 mentions, maintained a moderate positive correlation with sentiment (r = 0.31, p < 0.001).

The theoretical backbone of the study is Social Exchange Theory, the classic framework introduced by Peter Blau in 1964, which holds that workplace relationships operate on reciprocity: employees weigh the benefits they receive against the effort and commitment they contribute. When the exchange feels fair, workers respond with engagement, loyalty, and discretionary effort; when they feel undervalued, dissatisfaction and turnover intentions follow. Viewed through this lens, the findings suggest that employees increasingly interpret non-monetary rewards—flexible schedules, development opportunities, and work-life balance initiatives—as signals of organizational commitment to their well-being, not merely as transactional extras. The observed migration of discourse away from traditional compensation and toward holistic well-being aligns with this relational interpretation and with prior survey-based evidence that workers increasingly prioritize intangible benefits over pay alone.

The authors are candid about the limitations of their approach. The single-platform focus on LinkedIn may miss perspectives prevalent elsewhere; the one-year observation window means the quarter-to-quarter thematic shifts should be read as within-year movements rather than stable longitudinal trends; the English-only filter introduces cultural and linguistic bias; and LinkedIn’s user base skews toward certain professional and demographic groups. The dataset itself, consisting of publicly available comments collected under applicable data protection rules, cannot be shared in raw form, though anonymized and aggregated data are available on reasonable request. Despite these constraints, the study establishes a validated, replicable framework for social media analytics in compensation research—one the authors say achieved robust overall performance (F1 = 0.83) across the pipeline.

The practical implications for employers are considerable. As organizations compete for talent in a tight labor market, the study suggests that total rewards strategies built around pay alone may be increasingly out of step with workforce expectations. Compensation professionals now have evidence that flexible work arrangements and health benefits generate the strongest positive sentiment, that sentiment varies meaningfully by industry, and that the timing of benefits announcements visibly moves the needle on employee discourse. The researchers call for cross-platform validation, longitudinal studies spanning multiple years, multilingual analysis capabilities, and the integration of demographic variables to capture preference variation across employee segments. They even point toward predictive models capable of anticipating emerging compensation trends before they fully materialize in public conversation. In an era when employee voice is amplified, searchable, and machine-readable, the silent signals of the workforce are silent no more—and organizations that learn to listen computationally may gain a decisive edge in designing rewards that resonate.

Beyond its immediate findings, the study sits within a broader methodological turn in organizational research. Computational text analysis has gained traction across management science because it captures behavior in natural settings, sidestepping the artificiality of laboratory tasks and the recall problems of retrospective questionnaires. The choice of BERT is significant in this respect: unlike earlier bag-of-words techniques that ignore word order, transformer models process each word in relation to its surrounding context, allowing them to distinguish, for example, a sarcastic complaint about a benefits package from a sincere endorsement using nearly identical vocabulary. This contextual sensitivity matters greatly in compensation discourse, where negation, hedging, and irony are common.

The domain-specific adaptations the researchers made also illustrate a key lesson for applied text analytics. Off-the-shelf sentiment tools are typically trained on general web text or product reviews, where the language of workplace compensation is underrepresented. By supplementing the model with a custom lexicon of compensation terms and fine-tuning on manually labeled comments, the team addressed the vocabulary gap that often degrades performance when general-purpose models are applied to specialized professional discourse. The reported gap between the BERT model and the lexicon-based VADER baseline underscores how much accuracy can be lost without such adaptation.

The theoretical framing also deserves emphasis. Social Exchange Theory, as elaborated by scholars such as Gould-Williams and Davies, holds that employees interpret rewards not merely as transactional payments but as signals of how much the organization values them. The study’s finding that health insurance and flexible work arrangements carry the strongest positive sentiment fits this account: benefits that touch on security and personal autonomy may function as especially potent signals of organizational care. Conversely, the decline in base-pay discussion suggests that salary, while foundational, may be increasingly treated as a baseline expectation rather than a differentiator among employers.

For researchers, the study also highlights unresolved measurement questions. Because sentiment scores were aggregated across comments nested within users, posts, and organizations, the effective sample size for industry comparisons is smaller than the raw comment count suggests, and future multilevel designs could partition variance at each level. Extending the framework to multilingual corpora would be particularly valuable given that compensation norms and benefit expectations vary substantially across national labor markets. If subsequent work confirms these patterns across platforms and languages, social media analytics could become a routine complement to engagement surveys, giving organizations a near real-time barometer of how their reward strategies are actually landing with the workforce.

Subject of Research: Natural language processing analysis of employee perceptions of total rewards through LinkedIn discourse

Article Title: Employee perceptions of total rewards revealed through natural language processing of LinkedIn discourse

Article References: Altindağ, E., & Gül, Ö. (2026). Employee perceptions of total rewards revealed through natural language processing of LinkedIn discourse. Discover Global Society, 4(1), Article 238. https://doi.org/10.1007/s44282-026-00565-6

Image Credits: AI Generated

DOI: 10.1007/s44282-026-00565-6

Keywords: natural language processing, total rewards, LinkedIn, sentiment analysis, employee benefits, compensation, social media analytics, BERT, topic modeling, named entity recognition, work-life balance, human resource management

The mixed-methods study was conducted at NPHA between February and August 2025, three years after the agency was established under Sierra Leone’s National Public Health Agency Act. Researchers invited all 214 employees to complete a structured readiness survey, and 180 responded, an 84.1 percent response rate. The survey assessed 11 domains on a five-point scale, including surveillance, laboratory systems, information management, risk communication, emergency operations, logistics, emergency finance, coordination and two distinct workforce areas: internal human-resource management and outbreak surge capacity. The researchers also conducted 28 interviews with key informants and four focus-group discussions involving 32 participants. Staff were selected across senior leadership, technical and programme roles, operations and finance, and district-facing positions. Finally, the team reviewed institutional and operational records covering a 12-month period. This convergent design allowed perceived readiness to be compared with administrative evidence and with descriptions of how decisions, personnel, money and supplies moved through the emergency system.

Overall perceived readiness averaged 2.8 out of 5, with a standard deviation of 0.7; only 30.4 percent of respondents rated overall readiness at four or five. Internal human-resource management received the lowest score, averaging 2.1, and just 12.8 percent of staff gave it a high rating. By contrast, coordination with partners and the Ministry of Health averaged 3.6, surveillance and early warning 3.4, and risk communication 3.1. Workforce and surge capacity averaged 3.0, placing it above internal human-resource management. The distinction between these domains was central to the analysis. Surge capacity refers to the people and arrangements available to expand response during an outbreak, while internal human-resource management includes recruitment, induction, performance appraisal, career development, retention, payroll integration and routine personnel information systems. An agency may therefore possess trained responders and still lack the organisational machinery needed to recruit replacements, document responsibilities, support professional development or retain experienced staff after a crisis has passed.

Administrative records reinforced the survey results. NPHA had 214 employees in post against an approved establishment of 286 positions, an overall vacancy rate of 25.2 percent. The gaps were particularly pronounced in functions exposed to operational delays: finance and procurement positions were 60.0 percent vacant, district-facing posts 51.4 percent vacant, and human-resource and administration posts 44.4 percent vacant. Technical and programme positions were comparatively better filled, at 82.2 percent. Among 41 people appointed during the preceding year, only 13 had a documented induction record. The median interval from vacancy approval to appointment was 118 days, while only 13.1 percent of all staff had a completed annual appraisal on file. Updated job descriptions were present in 42.5 percent of personnel files, and documented continuing professional development during the previous year was available for 31.8 percent of staff. These figures suggest that individual expertise has developed faster than the systems needed to maintain and distribute it.

The agency’s response workforce is a significant asset, shaped by investments made after the 2014–2016 Ebola epidemic. Sierra Leone has developed a Field Epidemiology Training Programme, emergency operations infrastructure, electronic disease surveillance and response, laboratory resources and national and district Rapid Response Teams. Staff also carry experience from Ebola, COVID-19 and successive mpox outbreaks. The study found, however, that this strength remains vulnerable because much of it depends on a small, experienced cohort and informal personal networks. A national surge list included 74 staff, but it had not been updated for nine months; documented emergency roles or terms of reference existed for only 43 of those people. Just 26 had taken part in a simulation or drill during the preceding year. At district level, nine of 16 districts had updated Rapid Response Team lists, and five had conducted a documented simulation. The researchers interpret surge capacity as real but only partly institutionalised: staff know how to respond, yet the system does not consistently record who is responsible, how roles should be activated or how readiness should be rehearsed.

This concentration of capability was also visible in the statistical analysis. Technical staff with previous outbreak-response experience had substantially higher odds of reporting strong readiness than would be expected from the separate effects of technical training and experience alone. The adjusted odds ratio for their combined profile was 3.41, with a 95 percent confidence interval from 1.62 to 7.18. In practical terms, experienced technical personnel may be compensating for weak organisational systems through knowledge, relationships and the ability to improvise under pressure. That is useful during an emergency but creates institutional risk if those individuals leave. Of the 24 staff who exited during the preceding year, 15 were technical or programme staff, and 11 departures were recorded as movement to nongovernmental organisations, donor-supported projects or international agencies. Partner- or project-supported employees represented 40.7 percent of the workforce. Researchers say parallel employment arrangements can strengthen short-term capacity while complicating retention, reporting lines, career progression and the development of a unified agency identity.

Finance, procurement and decentralisation created additional delays between recognising an event and acting on it. Emergency financial systems averaged 2.3 out of 5, while logistics and supply-chain readiness averaged 2.4. The median time from approval of an activity to release of funds was 21 days, and emergency procurement took a median of 46 days. Sixteen of 43 reviewed emergency procurements, or 37.2 percent, exceeded their planned timelines. Records also documented seven episodes of reagent or personal protective equipment stockouts, while 18 staff advances required a median of 36 days for reimbursement. The agency’s national coordination was stronger than its documented district reach: only eight districts had designated NPHA focal persons, six had records of surveillance feedback to district teams, and deployment from a district request took a median of three days, increasing to five days for remote districts. These bottlenecks matter because outbreaks begin in communities and districts, whereas authority, financing and much of the coordination remain concentrated in Freetown.

The clearest operational signal came from examining the 7-1-7 framework, which separates detection within seven days of emergence, notification within one day of detection and completion of early response within seven days of notification. Across 12 priority events, the median time from emergence to detection was five days and from detection to notification was one day. Yet the median interval from notification to completion of early response was 18 days. Eight events met the detection target, nine met the notification target and only four completed early response within seven days. Delays were most often associated with specimen transport, laboratory confirmation, release of funds, field deployment, supplies and district logistics. The agency could therefore see and communicate public-health signals faster than it could turn them into completed field interventions. The study’s authors argue that this sequence reveals why legal authority, trained personnel and strong partnerships cannot by themselves demonstrate preparedness. For Sierra Leone and similar emerging public health institutes, durable readiness will depend on building integrated human-resource systems, documented surge roles, emergency financing, faster procurement and district routines that continue to function when projects end or experienced responders move on.

The findings also clarify how preparedness should be measured. A high-level assessment may confirm that surveillance, trained personnel or partner coordination exists, while missing the organisational steps that allow those resources to be activated under pressure. Reviewing personnel files, deployment records, procurement timelines and event histories alongside staff accounts therefore provides a more operational test of readiness. In this case, the agreement between perceptions and routine records strengthens the interpretation that administrative systems were not merely viewed unfavourably by employees; they were producing observable gaps in how the agency functioned.

The interaction between technical expertise and outbreak experience is particularly important for interpreting the regression result. The association does not show that either characteristic causes readiness, nor that experienced technical staff can substitute indefinitely for institutional systems. Rather, it suggests that expertise and practical exposure may reinforce one another, enabling a small group to navigate procedures, contacts and decisions more effectively. That pattern can make an organisation appear more capable during familiar emergencies while leaving it exposed to turnover, simultaneous events or threats outside the experience of its established responders. Readiness testing should consequently examine whether procedures work for less experienced staff, not only whether highly experienced personnel can deliver results.

For emerging public health institutes, the practical implication is to treat routine administration as part of the response architecture. Clear job descriptions, induction, appraisal, maintained rosters, delegated authority and documented district links create the conditions for expertise to be transferred and repeatedly used. These measures are less visible than laboratories or emergency operations centres, but they determine whether those assets can be connected into a timely response. Because this study examined one national agency over a defined period and used perceived readiness as one component of its analysis, its associations should not be treated as universal effect estimates. Its value lies instead in identifying testable institutional mechanisms that comparable agencies can examine through their own records, event timelines and role-specific accounts.

Subject of Research: Sierra Leone’s national public health emergency management capacity and operational capability

Article Title: A study of public health emergency management capacity and capability of Sierra Leone’s national public health agency

Article References: Ikoona, E. N., Namulemo, L., Sinnah, M. M., Vandi, M. A., & Sahr, F. (2026). A study of public health emergency management capacity and capability of Sierra Leone’s national public health agency. Journal of Emergency and Disaster Medicine, 2(1), Article 16. https://doi.org/10.1007/s44467-026-00020-1

Image Credits: AI Generated

DOI: 10.1007/s44467-026-00020-1

Keywords: Sierra Leone, public health emergencies, outbreak response, national public health institutes, health workforce, emergency preparedness, human resource management, surveillance, health systems, public, health, emergency