Social media engagement and participation intention in humanitarian logistics contexts: the mediating role of trust

Humanitarian organizations operate under conditions in which speed, coordination, transparency and access to vulnerable communities are critical. Humanitarian logistics therefore plays a central role in relief effectiveness by coordinating the flow of materials, information and resources across crisis-response networks (Altay et al., 2023). Improving logistical efficiency is therefore not a secondary operational concern but a decisive factor in reducing human suffering and strengthening community resilience.

At the same time, social media has become an important communication infrastructure for humanitarian organizations. Digital platforms allow organizations to disseminate urgent information, engage volunteers and potential donors, document field conditions and improve the visibility of relief activities. Interactive practices such as commenting, sharing and reposting can also create opportunities for public dialogue and wider participation in humanitarian campaigns (Iefimova and Pashchenko, 2025). As a result, digital interaction has become part of the broader humanitarian communication ecosystem, contributing to information exchange, awareness building and support for relief efforts (Erokhin and Komendantova, 2024a; Vhikai et al., 2024).

To better understand how such digital interaction influences humanitarian engagement, this study draws on the Stimulus–Organism–Response (S–O–R) theory as its overarching theoretical lens (Jacoby, 2002). The theory proposes that external environmental stimuli shape individuals’ internal evaluations and emotional states, which in turn influence their behavioral responses. In the context of this study, social media engagement and perceived logistics efficiency are conceptualized as two external stimuli that may be associated with participation intention. Trust represents the internal organism state through which these external cues may be translated into behavioral response, while participation intention constitutes the response itself. In this way, the S–O–R framework provides a coherent basis for explaining how digital engagement and perceived operational capability may be associated with intended humanitarian participation. Although perceived logistics efficiency is measured through respondents’ evaluations, it is modeled here as an external stimulus because the object of evaluation is an organizational capability, namely, the perceived efficiency of humanitarian logistics operations. In contrast, trust is modeled as the internal organism state because it reflects the respondent’s psychological confidence in the organization.

Within this process, the perception of humanitarian logistics efficiency occupies a central position. Affected communities and the broader public often assess humanitarian organizations based on their speed of response, fairness in resource distribution and transparency in supply chain management during crises (Altay et al., 2023; Schiffling and Piecyk, 2014). The more positively individuals perceive the efficiency of logistical operations, the more likely they are to develop confidence in the organization’s ability to deliver assistance effectively. Recent research suggests that social media strengthens this perception by disseminating field reports, visual evidence and real-time updates that allow audiences to observe how relief resources are managed on the ground, especially during critical phases of emergency response (Erokhin and Komendantova, 2024b; Lough, 2022).

Trust, likewise, remains one of the fundamental pillars of humanitarian action. Donors, beneficiaries and volunteers are unlikely to engage meaningfully with humanitarian organizations unless they perceive them as credible, transparent and reliable. Trust is developed gradually through the consistent provision of accurate information, visible documentation of activities and clear communication about how resources and donations are used (Snavely and Tracy, 2002). In this regard, social media has expanded the capacity of humanitarian organizations to build trust by enabling direct and immediate communication with the public, reducing informational distance and increasing operational transparency through real-time reporting from crisis zones (Lovejoy and Saxton, 2012; Fischer-Preßler et al., 2023).

From this perspective, the present study seeks to examine how social media engagement and perceived logistics efficiency shape individuals’ intention to participate in humanitarian activities within the Egyptian context, where humanitarian response is often shaped by multiple logistical and operational challenges. More specifically, the study examines the direct associations of social media engagement and perceived logistics efficiency with participation intention, as well as the mediating role of trust. The adoption of the S–O–R theory is especially important here, as it provides a structured explanation of how external organizational cues are translated into internal evaluation and, ultimately, into participation intention.

Prior research has examined important but largely separate components of this problem. Lovejoy and Saxton (2012) explain how nonprofit organizations use social media for information, community building and action, while Elvira-Lorilla et al. (2024) examine the role of social media in nonprofit fundraising; neither study integrates public perceptions of humanitarian logistics efficiency into its explanatory model. Conversely, Vhikai et al. (2024) examine digitalization and efficiency in humanitarian logistics, while Altay et al. (2023) review innovation in humanitarian logistics and supply chain management; these studies do not test how social media engagement and operational perceptions jointly shape public trust and participation intention. The literature therefore lacks an integrated explanation of how communication-based and operation-based cues combine to influence intended humanitarian support. This study addresses that gap by positioning trust as the internal evaluative mechanism connecting both types of cues to participation intention.

Accordingly, this study addresses the following research questions:

This study contributes to humanitarian logistics and nonprofit participation research in three specific ways. First, it shifts attention from internal humanitarian logistics performance to public perceptions of logistics efficiency as an external organizational cue that may shape trust and participation intention. Second, it integrates communication-based and operation-based stimuli within a single S–O–R model, allowing comparison between relational digital engagement and perceived operational competence. Third, it refines the S–O–R framework by showing that stimuli do not necessarily operate with equal behavioral strength. Social media engagement emerged as the strongest predictor of participation intention, whereas perceived logistics efficiency was strongly associated with trust but did not directly predict participation intention. This suggests that operational credibility may relate to participation intention mainly through trust, whereas interactive communication may mobilize participation largely independently of trust (Jacoby, 2002; Fischer-Preßler et al., 2023; Vieira, 2013; Chari and Ngcamu, 2025).

The conceptual foundation of this research is anchored in the S–O–R framework, which explains how external environmental cues shape individuals’ internal evaluations and, ultimately, their behavioral intentions. In the present study, social media engagement and perceived logistics efficiency are treated as stimulus variables because they represent external cues through which individuals form impressions of humanitarian organizations. Social media engagement reflects digital interaction with humanitarian content through viewing, sharing, commenting and responding to organizational updates, while perceived logistics efficiency reflects stakeholders’ evaluation of how effectively humanitarian organizations manage and deliver aid. Trust is treated as the organism variable because it represents the internal evaluative state developed in response to these cues, while participation intention is treated as the response, reflecting individuals’ willingness to donate, volunteer, advocate or share humanitarian campaigns.

Although perceived logistics efficiency is perceptual, it is modeled as a stimulus because respondents evaluate an external organizational capability: the visible efficiency of humanitarian logistics operations. Trust, by contrast, represents the respondent’s internal psychological confidence in the organization. This distinction is important because communication-based stimuli and operation-based stimuli may not activate participation intention in the same way. Social media engagement may create relational closeness, visibility and emotional relevance, whereas perceived logistics efficiency may mainly signal competence, reliability and accountability. Therefore, the model does not assume that all stimuli have equal behavioral strength.

Because S–O–R is primarily a process framework, this study also draws on trust theory and social exchange logic to explain why the proposed paths are expected to occur. Multidimensional trust theory suggests that perceived trustworthiness is shaped by assessments of organizational ability, benevolence and integrity (Mayer et al., 1995). Perceived logistics efficiency is expected to signal organizational ability, while social media engagement may communicate responsiveness, transparency and relational concern. Social exchange theory further suggests that individuals may reciprocate beneficial, responsive and credible interactions by developing stronger willingness to support the organization. In the present context, interactive humanitarian communication may create perceived relational value, while trust may reduce the uncertainty associated with donating, volunteering or advocating on the organization’s behalf. Accordingly, S–O–R provides the overall process structure, trust theory explains the internal evaluation of organizational ability, benevolence and integrity, and social exchange theory explains why positive organizational interactions may be reciprocated through supportive intentions (Ahmad et al., 2023; Cropanzano and Mitchell, 2005).

Although social media engagement is often assumed to encourage public participation, online interaction does not automatically translate into meaningful humanitarian support. In nonprofit and humanitarian contexts, users may like, share or comment on posts without donating, volunteering or providing sustained support. In addition, information overload, crisis fatigue, skepticism toward online campaigns and low perceived organizational credibility may weaken the relationship between digital engagement and participation intention. Therefore, the relationship between social media engagement and participation intention should be empirically tested rather than theoretically assumed (Fischer-Preßler et al., 2023).

Similarly, perceived logistics efficiency may not directly motivate participation intention. Although efficient logistics can signal competence and accountability, members of the public may perceive logistical performance as technical, distant or less emotionally salient than direct humanitarian stories and calls for support. Therefore, perceived logistics efficiency may mainly strengthen trust, while its direct association with participation intention may be weaker or context-dependent. This distinction strengthens the S–O–R logic by clarifying that not all stimuli are equally capable of activating behavioral responses (Jacoby, 2002; Vieira, 2013).

In recent years, there has been increasing interest in the role of social media interaction in humanitarian operations, as digital platforms have become important tools for information sharing, public engagement and communication with beneficiaries and potential supporters. Social media interaction including liking, sharing, commenting and reposting can increase the visibility of humanitarian needs and create opportunities for publics to engage with organizational campaigns (Lovejoy and Saxton, 2012; Fischer-Preßler et al., 2023).

The engagement literature commonly conceptualizes engagement as a multidimensional construct comprising cognitive, emotional and behavioral components (Brodie et al., 2011; Hollebeek et al., 2014). Together, these dimensions provide a multilayered understanding of how individuals connect with organizational content, making engagement a powerful predictor of behavioral responses. In the context of nonprofit and humanitarian organizations, social media serves as a primary channel through which these engagement dimensions are activated and translated into public involvement. Nonprofit organizations increasingly use social media to engage publics, strengthen stakeholder relationships and support fundraising and charitable action (Elvira-Lorilla et al., 2024).

Empirical evidence further reinforces this role, as social media enables wider visibility, rapid information exchange and collective mobilization, which together encourage the public to participate in humanitarian activities by sharing information, supporting initiatives and engaging in community-driven efforts (Lough, 2022; Iefimova and Pashchenko, 2025). Prior nonprofit and humanitarian communication research suggests that interactive social media can support public engagement by increasing visibility, dialogue and opportunities for stakeholders to share, respond to and participate in organizational campaigns (Lovejoy and Saxton, 2012; Fischer-Preßler et al., 2023). The availability of accurate information about response operations and transparent reporting of humanitarian activities may strengthen individuals’ willingness to support organizations during crises (Lough, 2022; Vhikai et al., 2024). Transparent and responsive online communication may also strengthen public trust by reducing information asymmetry and making organizational actions more visible to potential supporters (Lovejoy and Saxton, 2012; Fischer-Preßler et al., 2023). From a social exchange perspective, responsive and transparent online interaction may be experienced as a relational benefit that individuals reciprocate through willingness to support the organization, providing a direct behavioral route that does not depend on a prior trust judgment (Cropanzano and Mitchell, 2005). Therefore, drawing on the relational, motivational and communicative role of social media engagement, higher engagement with humanitarian organizations’ online content is expected to be positively associated with participation intention and trust. However, in line with the mixed evidence discussed earlier, this relationship should be empirically tested rather than assumed, because online engagement may not always translate into actual or intended support:

Perceived logistics efficiency refers to stakeholders’ evaluation of how effectively humanitarian organizations manage relief operations, including response speed, distribution accuracy, coordination, transparency and appropriate use of resources. In humanitarian contexts, logistics efficiency is not only an operational matter but also a public-facing signal of competence and accountability. When stakeholders perceive that aid is delivered efficiently and resources are managed responsibly, they may develop stronger confidence in the organization’s ability to fulfil its mission (Vhikai et al., 2024; Altay et al., 2023).

Although direct empirical evidence linking perceived logistics efficiency and trust in humanitarian contexts is limited, related literature provides a strong conceptual basis for expecting this relationship. Research in logistics service quality and humanitarian logistics suggests that reliability, transparency and operational performance may strengthen stakeholder confidence in an organization’s competence and accountability (Vhikai et al., 2024; Ali et al., 2022; Mentzer et al., 2001).

In humanitarian operations, where accountability, transparency and responsible resource allocation are critical, perceived logistics efficiency may therefore serve as a performance-based cue associated with trust formation, such that stronger perceptions of logistics efficiency are expected to be associated with higher trust in the organization.

From a behavioral perspective, a positive direct association between perceived logistics efficiency and participation intention is expected, because individuals may be more willing to support humanitarian organizations when they believe that their donations, volunteering efforts or advocacy will be converted into timely, efficient and meaningful humanitarian outcomes. This relationship is nevertheless theorized as weaker than the corresponding social media engagement pathway, because operational capability may first need to be translated into trust before it becomes behaviorally motivating. Therefore, the following hypotheses are proposed:

Participation intention reflects individuals’ willingness to support humanitarian activities through donating, volunteering, advocacy or sharing campaign-related information. In nonprofit and humanitarian contexts, these intentions are closely linked to trust because supporters often have limited direct oversight over how resources are collected, allocated and delivered. Trust reduces uncertainty and strengthens confidence that the organization will use resources responsibly and effectively (Snavely and Tracy, 2002). Research on charitable giving and nonprofit support similarly identifies organizational credibility, donor confidence and perceived effectiveness as important conditions for supportive intentions (Bekkers and Wiepking, 2011; Elvira-Lorilla et al., 2024):

Trust may operate as an internal evaluative mechanism that links external organizational cues to participation intention. Given the limited direct oversight described above, stakeholders rely on judgments about organizational ability, benevolence and integrity when deciding whether an organization merits support (Snavely and Tracy, 2002; Mayer et al., 1995). This logic is consistent with multidimensional trust theory, which conceptualizes trustworthiness through ability, benevolence and integrity. Perceived logistics efficiency mainly reflects the ability dimension because it signals whether the organization is competent in delivering aid effectively. Social media engagement, however, may also communicate benevolence and integrity by showing responsiveness, openness and concern for beneficiaries and supporters. This distinction helps explain why communication-based cues may be more effective in activating participation intention than operational cues alone.

Within the present model, trust is proposed as the mediating mechanism linking social media engagement and perceived logistics efficiency to participation intention. Social media engagement may strengthen trust by increasing transparency, interaction and organizational visibility, while perceived logistics efficiency may strengthen trust by signaling competence, responsiveness and responsible resource management. In turn, trust may increase individuals’ willingness to support humanitarian action. This mediation logic is consistent with the S–O–R framework and with mediation-based approaches that examine how external cues are translated into behavioral intentions through internal psychological mechanisms (Jacoby, 2002; Preacher and Hayes, 2008):

Based on the literature review, social media engagement and perceived logistics efficiency are proposed as external cues that may shape participation intention directly and indirectly through trust. Accordingly, Figure 1 presents the proposed research model and the hypothesized relationships among the study variables.

This study used a quantitative, cross-sectional survey design. Given the cross-sectional design, the study does not claim causal relationships among the constructs. Rather, the proposed paths represent theory-driven associations tested using structural equation modeling (SEM). Although the S–O–R framework supports the proposed direction of relationships, longitudinal or experimental studies would be required to establish causal ordering more conclusively.

Data were collected using an online questionnaire distributed to Egyptian social media users who had prior exposure to humanitarian organizations’ content (e.g. donation campaigns, emergency appeals, field updates or volunteering announcements). The target population included adults residing in Egypt who actively use social media platforms and are potential donors and/or volunteers for humanitarian initiatives.

Table 1 presents a summary of the demographic characteristics of the respondents. The table reports the distribution of age, gender and educational level and average time spent on social media. The sample included respondents from different gender, age, educational and social media usage groups. However, because the study used non-probability convenience and purposive sampling, the sample should be considered heterogeneous rather than statistically representative of the wider Egyptian population.

The final sample (n = 574) comprised 274 male respondents (47.74%) and 300 female respondents (52.26%). The age distribution was 68 respondents aged 18–24 (11.85%), 162 aged 25–34 (28.22%), 172 aged 35–44 (29.97%), 123 aged 45–55 (21.43%) and 49 aged over 55 (8.54%). Educational levels comprised 27 undergraduates (4.70%), 61 diploma holders (10.63%), 374 bachelor’s degree holders (65.16%), 74 master’s degree holders (12.89%) and 38 PhD holders (6.62%). Daily social-media use was less than one hour for 11 respondents (1.92%), one to three hours for 117 respondents (20.38%), three to five hours for 355 respondents (61.85%) and more than five hours for 91 respondents (15.85%).

The use of convenience and purposive sampling was appropriate because the study required respondents who had prior exposure to humanitarian organizations’ social media content. Nevertheless, this approach may introduce self-selection bias, as individuals who are more interested in humanitarian causes or more active on social media may have been more likely to participate. Therefore, the findings should be generalized with caution.

All constructs were measured using multi-item scales adapted from previous validated literature, with wording modified to fit the humanitarian context in Egypt. A five-point Likert scale was used for all items, ranging from 1 = “strongly disagree” to 5 = “strongly agree.” The questionnaire was developed in a clear and concise format to support respondent understanding.

Social media engagement was measured using eight items adapted from Iliopoulou et al. (2024) based on the social media engagement framework of Chu et al. (2020). Perceived humanitarian logistics efficiency was measured using four items adapted from Masudin et al. (2020) and framed from a customer-oriented perspective (Schiffling and Piecyk, 2014). Trust in humanitarian organizations was measured using four items adapted to the humanitarian context from Hou et al. (2020). The items assessed whether humanitarian organizations act in the interests of crisis-affected people, operate ethically, keep their promises and are trusted overall. Finally, participation intention was measured using four items adapted from Ahn and Lee (2019), capturing individuals’ willingness and likelihood to participate in future humanitarian relief efforts. The complete questionnaire items and their original sources are presented in  Appendix 1.

Following the development of the questionnaire, the instrument was translated into Arabic and then verified using the back-translation technique to ensure linguistic accuracy and conceptual equivalence between the original and translated versions (Klotz et al., 2023). The final Arabic survey was administered online using Google Forms. To obtain respondents who were relevant to the study context, a non-probability sampling strategy combining convenience sampling and purposive screening was used. Specifically, eligibility was limited to individuals who reported following or engaging with humanitarian-related pages, posts or campaign content on social media. The survey link was distributed openly across several widely used social networking platforms, including LinkedIn, Facebook, Instagram, WhatsApp and Messenger. Participation was voluntary, and responses were collected anonymously. This process resulted in a total of 574 valid responses. Because the survey was disseminated openly through social media rather than to a predefined list of recipients, a conventional response rate could not be calculated (Shiyab et al., 2023).

Prior to the full administration of the survey, a pilot study was conducted to validate the survey’s objectives and content. A total of 44 responses were collected to assess the instrument’s clarity, validity and contextual relevance, and minor revisions were subsequently made based on the feedback received. To further strengthen the instrument, both face validity and content validity were established through expert review involving four academics and five practitioners. This process helped ensure that the questionnaire items were clear, understandable and appropriately aligned with the constructs being measured.

Data analysis was conducted in two stages. First, the measurement model was evaluated through confirmatory factor analysis to assess reliability and validity. Second, the structural model was tested using SEM with maximum likelihood estimation in IBM AMOS. The measurement model results are presented in Table 2.

The measurement model showed acceptable reliability and convergent validity. All standardized factor loadings exceeded the acceptable threshold of 0.60. Composite reliability values ranged from 0.858–0.895, exceeding the recommended threshold of 0.70. Average variance extracted (AVE) values ranged from 0.518–0.638, exceeding the minimum threshold of 0.50. These results support internal consistency and convergent validity (Fornell and Larcker, 1981). The measurement model was specified with each of the 20 indicators loading on a single construct; no cross-loadings were freed. Based on modification indices and theoretical justification, seven error covariances were freed between adjacent items measured within the same construct: SME2–SME3, SME4–SME5, SME6–SME7, SME7–SME8, PLE1–PLE2, PLE3–PLE4 and PI1–PI2. Correlated errors of this type are considered acceptable when they occur between items belonging to the same construct and can be attributed to shared wording or item proximity in the questionnaire, rather than across constructs, where they would distort the estimated structural relationships (Byrne, 2016; Cole et al., 2007). Freeing these seven parameters reduced the degrees of freedom from 164, the value expected for a four-factor model with 20 indicators and no freed error covariances, to 157. Table 3 reports the fit indices for the final model (Model 2); for transparency, a full comparison with the corresponding model estimated without the freed error covariances (Model 1) is provided in  Appendix 2.

The model demonstrated an acceptable overall fit. The incremental fit indices were within acceptable ranges (NFI = 0.927, IFI = 0.947, TLI = 0.936, CFI = 0.947), and RMSEA indicated acceptable approximate fit (RMSEA = 0.064). The normed chi-square was also acceptable (CMIN/DF = 3.331). However, AGFI was slightly below the conventional 0.90 threshold, and RMR = 0.100 suggests some residual misfit. Accordingly, model fit was evaluated using the full pattern of fit indices rather than relying on a single index, and the RMR value is acknowledged as a limitation of the measurement model.

Freeing the seven within-construct error covariances produced a significant improvement over the corresponding model without them [chi-square difference test: Δχ²(7) = 242.96, p < 0.001; see  Appendix 2]. In Model 1, all standardized loadings remained above 0.60 and AVE exceeded 0.50 for all four constructs (social media engagement = 0.532, perceived logistics efficiency = 0.637, trust = 0.638, participation intention = 0.615), values closely comparable to those reported in Table 2. The freed error covariances were also modest in magnitude, with standardized values ranging from 0.167–0.426 in absolute terms. The standardized residual association between PLE3 and PLE4 was negative (−0.391), indicating an inverse association between their residuals after accounting for the common factor. This additional measurement complexity is acknowledged as a limitation.

Discriminant validity was then assessed using the Fornell and Larcker (1981) criterion. To ensure that this assessment was conducted at the latent rather than the composite-score level, both the AVE values and the inter-construct correlations were obtained from Model 1, in which all four constructs were specified as freely correlated factors with no directional paths and no freed error covariances. As shown in Table 4, the square root of AVE exceeded the corresponding latent correlation for every construct pair except perceived logistics efficiency and trust, where the latent correlation (r = 0.822) exceeded the square root of AVE for both constructs (0.798 and 0.799, respectively). The same conclusion is reached if the AVE values from the final model reported in Table 2 are used instead (0.791 and 0.799), indicating that this result is not an artefact of the model specification.

Discriminant validity was further assessed using the heterotrait–monotrait ratio of correlations based on the item-level correlation matrix. As shown in Table 5, the heterotrait–monotrait (HTMT) values ranged from 0.278–0.830, and all values were below the conservative threshold of 0.85 (Henseler et al., 2015). These findings provide additional evidence that the constructs are empirically distinguishable. The highest value was observed between perceived logistics efficiency and trust (HTMT = 0.830). Although this value remained below the threshold, its proximity to 0.85 indicates that the distinction between the two constructs should be interpreted cautiously and considered together with the Fornell–Larcker results above and the chi-square difference test reported below.

Because the Fornell–Larcker criterion did not support full discriminant validity between perceived logistics efficiency and trust, a formal chi-square difference test was conducted to determine whether the two constructs are empirically distinguishable (Anderson and Gerbing, 1988). Two nested models were compared, both estimated without freed error covariances: the four-factor model (Model 1) and a three-factor model in which the four perceived logistics efficiency items and the four trust items were specified as indicators of a single merged factor. As shown in Table 6, merging the two constructs produced a significant deterioration in fit relative to the four-factor model [Δχ²(3) = 191.18, p < 0.001], with corresponding declines in CFI (0.913–0.886) and TLI (0.899–0.870) and an increase in RMSEA (0.080–0.091). This result indicates that perceived logistics efficiency and trust cannot be represented by a single construct without a substantial loss of fit, supporting their treatment as empirically distinct constructs despite the strong latent correlation between them.

The elevated correlation is not unexpected. All four perceived logistics efficiency items and three of the four trust items share an identical stem (“I believe humanitarian organizations […]”), and two items in particular, TRU3 (“[…] keep their promises about humanitarian assistance”) and PLE2 (“[…] deliver humanitarian aid within the expected time”) are conceptually close. This shared item framing likely contributes to the strength of the observed association and is addressed further as a limitation below.

Common method bias was addressed procedurally through respondent anonymity, voluntary participation and clear item wording. Statistically, Harman’s single-factor test showed that the first unrotated factor explained 38.35% of the total variance, below the 50% threshold. A full-collinearity assessment also produced variance inflation factor (VIF) values of 1.64 for social media engagement, 2.11 for perceived logistics efficiency, 2.21 for trust and 1.73 for participation intention, all below 3.3 (Kock, 2015). These results suggest that common method bias was unlikely to dominate the observed relationships. Nevertheless, it cannot be completely ruled out because the study used cross-sectional self-reported data. Regarding nonresponse bias, Levene’s test was nonsignificant, indicating no significant difference in variance between early and late responses (Armstrong and Overton, 1977).

This section presents the results of the structural model and hypothesis testing. The analysis examined the direct associations among social media engagement, perceived logistics efficiency, trust and participation intention, as well as the mediating role of trust in the proposed model. Before testing the hypotheses, multicollinearity was assessed using tolerance and VIF values based on composite scores. For participation intention, social media engagement, perceived logistics efficiency and trust were entered simultaneously as predictors. The VIF values ranged from 1.096–2.162, while tolerance values ranged from 0.462–0.913. For trust, social media engagement and perceived logistics efficiency were entered simultaneously as predictors, with a VIF of 1.065 and a tolerance of 0.939 for both predictors. Since all VIF values were below the recommended threshold of 3.3 and all tolerance values exceeded 0.20, multicollinearity was not considered a serious concern.

The structural model and hypothesis-testing results are presented in Table 7. The table reports the standardized path coefficients for the direct relationships, the indirect effects for the mediation hypotheses, the confidence intervals for the indirect effects and the final decision for each hypothesis.

The structural results show that social media engagement was positively associated with participation intention (β = 0.653, p < 0.001), supporting H1. Social media engagement was also positively associated with trust (β = 0.133, p < 0.001), supporting H2. Perceived logistics efficiency showed a strong positive association with trust (β = 0.803, p < 0.001), supporting H3. However, perceived logistics efficiency was not significantly associated with participation intention (β = 0.009, p = 0.909), and H4 was therefore not supported. Trust was positively associated with participation intention (β = 0.187, p = 0.024), supporting H5.

Regarding the mediation hypotheses, inference was based on the 95% bias-corrected bootstrap confidence interval of the unstandardized indirect effect, with the percentile interval reported as a sensitivity check. The indirect association between social media engagement and participation intention through trust was small in magnitude (β = 0.025). Its bias-corrected interval excluded zero [95% CI (0.002, 0.062), p = 0.037], supporting H6, although the percentile interval marginally included zero [95% CI (−0.001, 0.059), p = 0.055].

The indirect association between perceived logistics efficiency and participation intention through trust was substantially larger (β = 0.150), but both the bias-corrected [95% CI (−0.001, 0.341), p = 0.051] and the percentile [95% CI (−0.003, 0.341), p = 0.055] intervals included zero; H7 was therefore not supported. This imprecision reflects mainly the trust → participation intention path, which was significant in the maximum-likelihood test used for the direct paths (p = 0.024) but whose bootstrap interval marginally included zero [95% bias-corrected CI (−0.003, 0.362), p = 0.053]. The model explained 70.9% of the variance in trust and 54.3% of the variance in participation intention.

The total association between perceived logistics efficiency and participation intention was statistically significant [b = 0.187, 95% bias-corrected CI (0.090, 0.299), p < 0.001; β = 0.159]. Expressed as a share of each variable’s total effect on participation intention, trust carried roughly 94% of the perceived-logistics-efficiency relationship (0.150 of 0.159) but only about 4% of the social-media-engagement relationship (0.025 of 0.678). Perceived logistics efficiency is therefore associated with participation intention predominantly through trust, although this indirect component could not be distinguished from zero at the 95% level, whereas social media engagement is associated with it largely independently of trust.

This study contributes to humanitarian communication and logistics research by showing that communication-based and operation-based stimuli differ in their association with trust and participation intention. Interpreted through the S–O–R framework, social media engagement and perceived logistics efficiency operate as external stimuli, trust represents the internal organism state and participation intention reflects the response. The findings indicate that the proposed mechanism is not fully symmetric: social media engagement is directly associated with participation intention and also operates indirectly through trust, whereas perceived logistics efficiency is strongly associated with trust but has no statistically significant direct association with participation intention. Because perceived logistics efficiency has no meaningful direct link to participation intention, trust carries most of its estimated association with the response, whereas social media engagement’s link to participation intention holds largely independent of trust.

The positive role of social media engagement in shaping participation intention is consistent with prior work arguing that digital interaction is not merely informational, but relational and mobilizing (Lovejoy and Saxton, 2012; Fischer-Preßler et al., 2023). Studies on engagement have long shown that cognitive, emotional and behavioral involvement with organizational content strengthens downstream behavioral responses (Barari et al., 2025; Hollebeek et al., 2014). In the nonprofit and humanitarian context, social media creates a participatory environment in which publics do not simply consume messages; rather, they observe, react, share and become socially embedded in the organization’s activities (Lovejoy and Saxton, 2012; Lough, 2022; Elvira-Lorilla et al., 2024). The present findings reinforce this view by suggesting that public willingness to support humanitarian action is particularly responsive to visible and dialogic forms of communication. In practical terms, this means that public participation appears to be driven less by static communication and more by continuous, interactive contact that gives humanitarian action social immediacy and relevance. This pattern is consistent with social exchange logic: engagement appears to be reciprocated directly through supportive intentions, rather than requiring conversion into a trust judgment first (Cropanzano and Mitchell, 2005).

The positive but modest association between social media engagement and trust indicates that digital communication contributes to trust formation, consistent with the view that responsiveness, transparency and real-time updates can reduce uncertainty and strengthen perceived legitimacy (Fischer-Preßler et al., 2023; Saxton and Wang, 2014). This trust-building role, however, appears secondary to engagement’s direct behavioral pull: social media engagement is associated with participation intention mainly through its direct effect, with trust carrying only a limited share of that relationship. From an S–O–R perspective, this suggests that a communication-based stimulus can mobilize the response largely without first being converted into the internal organism state a pattern consistent with the distinction that Mayer et al. (1995) draw between ability and the benevolence and integrity dimensions of trustworthiness: signals of responsiveness and goodwill may motivate participation more directly, whereas ability-based signals appear to operate mainly through trust.

The strong association between perceived logistics efficiency and trust provides an important operational insight. Humanitarian stakeholders appear to interpret efficiency as evidence of organizational ability, the ability dimension of trustworthiness set out in the framework above (Mayer et al., 1995). This finding agrees with the logistics and service-quality literature, which has consistently treated reliability, timeliness and performance competence as foundations of stakeholder confidence (Vhikai et al., 2024; Mentzer et al., 2001). It also resonates with humanitarian operations scholarship emphasizing that effective delivery systems signal discipline, coordination and responsible resource management (Altay et al., 2023). In this respect, the results position logistics efficiency as more than a technical function: it becomes an external credibility signal from which publics infer whether the organization is capable of fulfilling its mission (Vhikai et al., 2024).

The nonsignificant direct association between perceived logistics efficiency and participation intention provides an important qualification. Operational competence may strengthen confidence in a humanitarian organization without necessarily becoming an immediate behavioral trigger. Logistics performance is likely to be perceived as more technical and less emotionally salient than interactive messages, beneficiary narratives and direct calls to action. Within the S–O–R framework, perceived logistics efficiency may therefore operate mainly through internal evaluation, particularly trust, rather than directly activating participation intention. Operational quality may be important for credibility, but the present findings suggest that it is insufficient on its own to mobilize intended public support (Mehrabian and Russell, 1974; Jacoby, 2002; Mayer et al., 1995).

In charitable and humanitarian settings, this mechanism is especially salient because supporters rarely have direct oversight over the allocation of resources or the implementation of aid. As a result, trust functions as a psychological shortcut through which people infer that their contribution will be used appropriately and generate real impact. This interpretation is consistent with nonprofit and charitable giving research showing that donor confidence, trust and perceived organizational credibility are important drivers of supportive intentions and charitable support (Elvira-Lorilla et al., 2024; Saxton and Wang, 2014). Accordingly, the present study supports the view that trust is not a secondary attitudinal variable, but a core mechanism connecting organizational evaluation to behavioral readiness.

The mediation results are consistent with the S–O–R proposition that responses can be generated through internal organism states (Preacher and Hayes, 2008; Mehrabian and Russell, 1974), but the role of trust differs markedly between the two stimuli. Relative to each stimulus’s overall association with participation intention, trust accounts for nearly the entire perceived-logistics-efficiency pathway, since its direct route to participation intention is negligible, whereas it accounts for only a limited part of the social-media-engagement pathway, which is dominated by its direct effect. Because the social-media indirect effect was modest in absolute size and significant only under the bias-corrected interval, and the logistics-efficiency indirect effect could not be distinguished from zero at the 95% level, this asymmetry rests on the relative size of the estimates rather than on their statistical significance and should be interpreted cautiously rather than treated as a definitive boundary condition of the S–O–R framework.

Although the indirect effect of perceived logistics efficiency through trust was not statistically significant at the 95% level (H7), it accounted for approximately 94% of a significant total association, because the direct path was negligible. This pattern suggests that operational credibility is unlikely to mobilize participation intention on its own and that its association with intention runs predominantly through trust, a pathway that requires confirmation with more precise estimates. A plausible explanation is that logistics efficiency primarily signals organizational ability and competence, whereas participation intention may also require emotional connection, perceived care, moral credibility and social identification. These relational and affective cues are more likely to be conveyed through interactive communication, storytelling, public responsiveness and transparent field updates than through efficiency perceptions alone. This predominantly indirect pattern does not mean that logistics efficiency is unimportant; rather, its value for participation lies in the trust it builds.

Overall, the findings refine the S–O–R framework by showing that communication-based and operation-based stimuli differ in their ability to activate trust and participation intention. Social media engagement operates mainly as a direct mobilizing stimulus, with trust playing only a supporting role in that relationship. Perceived logistics efficiency, in contrast, appears to operate predominantly as a trust-building signal: with no meaningful direct link to participation intention, its association with the response appears to depend on first being converted into trust, although this indirect pathway was not statistically confirmed. This distinction provides a clearer theoretical explanation of why digital engagement can mobilize intended public support somewhat independently of a fully formed trust judgment, while operational credibility appears to require that conversion to become behaviorally relevant (Fischer-Preßler et al., 2023; Mehrabian and Russell, 1974).

This study contributes to the theoretical development of the S–O–R framework in humanitarian logistics and nonprofit participation research. Rather than treating all stimuli as equally influential, the study distinguishes between communication-based and operation-based stimuli and shows that they differ in their behavioral strength. Social media engagement showed the strongest direct association with participation intention.

The findings suggest that social media engagement functions mainly as a direct mobilizing stimulus, with trust carrying only a limited share of its influence, whereas perceived logistics efficiency appears to operate almost entirely as a trust-building cue whose estimated association with participation intention was predominantly indirect through trust. This distinction extends the S–O–R framework by showing that stimuli can differ not only in strength but also in their dependence on the organism state.

The findings also qualify the organism component of the S–O–R framework by suggesting that ability-based cues, represented by perceived logistics efficiency, may relate to participation intention mainly through trust, whereas cues that also communicate responsiveness, benevolence and integrity, as social media engagement appears to in this context can mobilize the response more directly, with less dependence on that conversion. This interpretation links the S–O–R process structure to multidimensional trust theory, although it remains tentative because the social-media indirect effect was small in magnitude and the perceived-logistics-efficiency indirect pathway was not statistically significant at the 95% level; it should therefore be treated as a proposition to be tested in future research. Social exchange theory offers a complementary explanation for the direct association between social media engagement and participation intention: responsive online interaction may be experienced as a relational benefit that individuals reciprocate through supportive intentions, although reciprocity was not measured separately in this study.

The findings offer several practical implications for humanitarian organizations. First, social media should be managed as a participation pathway rather than merely as a visibility tool. Humanitarian organizations should include clear calls to action in posts, such as donation links, volunteer registration forms, campaign-sharing options and emergency support instructions. Second, because social media engagement showed the strongest association with participation intention, organizations should prioritize interactive communication, including responding to public questions, publishing field updates and sharing evidence of aid delivery.

Third, although perceived logistics efficiency did not directly predict participation intention, it was strongly associated with trust. Therefore, logistics performance should be made visible through simple public-facing updates, such as delivery progress, distribution locations, response speed and resource-use summaries. Fourth, organizations should not evaluate digital communication only through likes, shares and reach; they should also track trust-related indicators such as perceived transparency, credibility and confidence in resource use. However, because the study measured participation intention rather than actual donation or volunteering behavior, these recommendations should be interpreted as strategies for strengthening willingness to participate rather than as direct evidence of increased participation behavior.

From a societal perspective, the findings suggest that transparent digital communication and visible logistics reporting can strengthen public confidence in humanitarian organizations. This is particularly important in crisis-prone environments, where public willingness to donate, volunteer or advocate may depend on whether organizations are perceived as credible, responsive and accountable. Humanitarian organizations and policymakers should therefore encourage communication practices that make aid delivery processes more transparent and understandable to the public.

Despite its contributions, this study is subject to several limitations. First, the cross-sectional design restricts the ability to infer causal relationships among the study variables. Future research could use longitudinal or experimental designs to examine causal ordering more rigorously. Second, the use of convenience and purposive sampling through open online distribution may introduce self-selection bias, as respondents with stronger interest in humanitarian issues or greater social media activity may have been more likely to participate. Future studies could use probability-based, stratified or organization-based sampling strategies where feasible.

Third, this study measured participation intention rather than actual donation, volunteering or advocacy behavior. Future research should examine actual behavioral data to assess whether intention translates into real humanitarian support. Fourth, the study was conducted in Egypt, which may limit the generalizability of the findings to other humanitarian and national contexts. Future studies could validate the model across different countries, crisis types and humanitarian organizations. Fifth, future research could examine additional organism variables, such as empathy, perceived transparency, moral identification, perceived urgency or organizational reputation, to better explain how humanitarian stimuli are translated into participation behavior. Sixth, the indirect effect of perceived logistics efficiency through trust could not be distinguished from zero at the 95% level, and that of social media engagement was significant only under the bias-corrected interval; larger samples would allow these indirect effects to be estimated more precisely.

Finally, two measurement-related limitations should be noted. Although the chi-square difference test supported the discriminant validity of perceived logistics efficiency and trust, the elevated latent correlation between them (r = 0.822), together with the unmet Fornell–Larcker criterion for this pair and the borderline HTMT value (0.830), indicates a close empirical and conceptual relationship. This may partly reflect the shared item wording used to measure both constructs, as most items for both began with the same stem and two items in particular (TRU3 and PLE2) overlap substantively in content. Relatedly, the measurement model required seven freed error covariances between adjacent items within the same constructs to achieve acceptable fit; although all were within-construct and theoretically justifiable on the basis of shared wording, their presence indicates some residual overlap in item content. Future research should therefore consider more clearly differentiated item wording, or complementary measurement approaches such as behavioral or third-party indicators of logistics performance, to further separate ability-based operational judgements from global organizational trust.

This study examined how social media engagement and perceived logistics efficiency are associated with participation intention in humanitarian logistics contexts, with trust positioned as the internal evaluative mechanism within the S–O–R framework. The findings show that social media engagement is strongly associated with participation intention mainly through a direct pathway, with trust carrying only a small share of that relationship. Perceived logistics efficiency, by contrast, is strongly associated with trust but not directly associated with participation intention; approximately 94% of its association with participation intention runs through trust, although this indirect effect was not statistically significant at the 95% level. These findings suggest that, in humanitarian settings, operational capability is unlikely to mobilize public support on its own: its association with participation intention appears to run mainly through trust, whereas interactive digital communication is associated with support more directly. Future research should validate these relationships using longitudinal, experimental or behavior-based designs.