Rural social cohesion dynamics and their influence on collective adoption of extension-promoted technologies

Introduction

A 2021 survey across West Africa found that 64% of smallholder farmers made technology adoption decisions based on what their neighbors did, not on extension recommendations alone [1] . This statistic exposes a blind spot in conventional extension thinking, which still largely treats adoption as an individual-level process driven by knowledge, resources, and attitudes [2] . In reality, rural communities across sub-Saharan Africa operate through dense webs of social obligation, shared labor, and collective risk management that shape how new practices spread or don't [3] . Social cohesion defined here as the strength of bonds, trust levels, and shared norms within a community has received growing attention in development research but remains underexplored in extension science [4] . Isham [5] showed that social capital predicted adoption of soil conservation in Tanzania more reliably than household wealth. Bandiera and Rasul [6] found that farmers with many social connections adopted new crops faster, but only up to a threshold beyond which conformity pressure slowed experimentation. These findings suggest the relationship between cohesion and adoption is not linear but shaped by community power dynamics and cultural norms. Liberia presents a distinctive context. Decades of civil conflict fragmented rural social structures, and post-war reconstruction has rebuilt community bonds unevenly across regions [7] . Extension services restarted in the mid-2000s but have struggled with variable reception across communities that appear similar in economic terms yet differ markedly in social organization [8] . This research examined how social cohesion dynamics influence collective technology adoption in Bong County, testing whether cohesion acts as a straightforward enabler or a more complex moderating force.

Theoretical Framework

The research drew on Putnam's [9] social capital theory, distinguishing between bonding capital (ties within groups) and bridging capital (ties across groups). We combined this with Rogers' [10] diffusion of innovations framework, particularly the concepts of opinion leadership and social system norms. The integrated model hypothesized that bonding capital would accelerate adoption within groups but potentially slow it when bridging capital was weak, creating insular decision-making. This dual-pathway framework allowed us to test both the enabling and constraining effects of cohesion on collective technology uptake.

Material and Methods

Material

This cross-sectional research was conducted in Bong County, Liberia (7.08°N, 9.36°W), through Cuttington University between April and October 2022. Bong County www.extensionjournal.com has an estimated rural population of 333,000 with agriculture supporting over 82% of livelihoods. Eighteen communities were selected through purposive sampling to represent high, moderate, and low social cohesion levels as assessed through preliminary key informant interviews. The sample comprised 296 farming households (proportionally allocated across cohesion categories), 24 focus group discussions (8 per cohesion level, sex-disaggregated), and 12 extension worker interviews. Social cohesion was measured using an adapted Neighborhood Cohesion Instrument covering trust, belonging, participation, and shared values on a 5-point Likert scale. Ethical approval was granted by the Cuttington University Institutional Review Board (CU-IRB-2022-031, approved March 8, 2022).

Methods

Quantitative analysis used SPSS version 26. Pearson correlations assessed bivariate relationships between cohesion indicators and adoption variables. Hierarchical multiple regression tested the predictive power of social cohesion after controlling for household demographics and economic factors. Binary logistic regression identified predictors of collective (versus individual) adoption decisions. Focus group transcripts were analyzed through directed content analysis using pre-defined categories from the theoretical framework, with emergent themes added iteratively. Social network density was calculated for each community using reported information-sharing ties. All quantitative tests used p< 0.05 significance threshold.

Results

Social cohesion variables (

Step 2) explained an additional 38.4% of variance beyond demographics (Table 2). Community trust was the strongest predictor (β = 0.39), followed by information network density (β = 0.28) and participation norms (β = 0.19). The full model explained 47.3% of total variance (F = 26.8, p< 0.001). Most communities fell in the moderate cohesion category (38.6%), while only 27.3% scored as high cohesion (Figure 1). The 11.3% classified as very low cohesion were predominantly communities with recent internal displacement or land disputes. The pathway model (Figure 2) shows that social cohesion influenced adoption through two parallel channels: trust in extension agents (r = 0.61) and community participation leading to information-sharing networks (r = 0.58). Both pathways converged on collective adoption, with moderating effects from gender norms, land tenure, and market access.

Discussion

The 2.7-fold difference in collective adoption between high and low-cohesion communities confirms that social bonds act as a powerful channel for extension impact-but not an uncomplicated one. The finding that cohesion variables explained 38.4% of adoption variance after controlling for demographics places social architecture alongside economic factors as a primary driver of technology spread [11] . This aligns with Wossen et al. [12] , who reported that social networks predicted improved maize variety adoption in Ethiopia independently of wealth indicators. The conformity effect observed in 23.1% of cohesive communities warrants careful attention. When influential community members opposed a promoted technology, tight social bonds amplified that resistance rather than buffering it. This mirrors what Bandiera and Rasul [6] termed "negative social effects" and suggests that extension agents need to identify and engage opinion leaders early in the diffusion process [13] . Focus group data revealed that in several high-cohesion communities, a single elder's skepticism about a new seed variety delayed group adoption by over eight months. Gender dynamics within cohesive communities also shaped outcomes asymmetrically. Women in highly cohesive communities reported stronger pressure to conform to collective decisions, even when they individually favored adopting a promoted technology. This finding complicates the assumption that cohesion uniformly benefits all community members and supports Meinzen-Dick et al.'s [14] argument that social capital operates differently across gender lines.

Limitations

The purposive selection of communities based on preassessed cohesion levels may have introduced selection bias. Social cohesion was measured at a single time point, missing temporal dynamics that shape adoption over seasons. Self-reported trust and participation data are vulnerable to social desirability bias. The research covered one county in Liberia, and post-conflict social dynamics may limit generalizability to other West African settings. Longitudinal research tracking cohesion changes alongside adoption trajectories would strengthen causal claims.

Conclusion

Social cohesion powerfully shapes how farming communities respond to extension-promoted technologies, but the relationship is more complex than a simple "more cohesion equals more adoption" formula. High-cohesion communities adopted collectively at nearly three times the rate of fragmented ones, and social variables outweighed demographic predictors. Yet the conformity effect and gender asymmetries found here indicate that cohesion can also constrain individual agency and slow adoption when opinion leaders resist change. Extension programs operating in post-conflict settings like Liberia need to map community social structures before designing delivery strategies. Working through existing trust networks appears more effective than bypassing them, but agents must also identify potential resistance nodes and engage them proactively. Gender-sensitive approaches that create space for women's independent decision-making within collective frameworks are equally important. The practical recommendation is to add social cohesion assessment to the standard extension planning toolkit alongside economic and agronomic surveys.

Acknowledgement

Acknowledgements Funding SourcesThis research was funded through an institutional research allocation from Cuttington University. No external

References

1. Mwangi M, Kariuki S. Factors determining adoption of new agricultural technology by smallholder farmers in developing countries. Journal of Economics and Sustainable Development. 2015;6(5):208-216.
2. Maertens A, Barrett CB. Measuring social networks' effects on agricultural technology adoption. American Journal of Agricultural Economics. 2013;95(2):353359.
3. Liverpool-Tasie LSO, Winter-Nelson A. Social learning and farm technology in Ethiopia: impacts by technology and adopter type. Journal of Development Studies. 2018;54(10):1689-1708.
4. Cramb RA. Processes affecting the successful adoption of new technologies by smallholders. Canberra: Australian Centre for International Agricultural Research; 2020. (ACIAR Working Paper Series No. 62).
5. Isham J. The effect of social capital on fertiliser adoption: evidence from rural Tanzania. Journal of African Economies. 2002;11(1):39-60.
6. Bandiera O, Rasul I. Social networks and technology adoption in northern Mozambique. Economic Journal. 2006;116(514):869-902.
7. Richards P. Fighting for the rain forest: war, youth and resources in Sierra Leone. Oxford: James Currey Publishers; 2006.
8. Tsai W, Ghoshal S. Social capital and value creation: the role of intrafirm networks. Academy of Management Journal. 1998;41(4):464-476.
9. Putnam RD. Bowling alone: the collapse and revival of American community. New York: Simon and Schuster; 2000.
10. Rogers EM. Diffusion of innovations. 5th ed. New York: Free Press; 2003.
11. Genius M, Koundouri P, Nauges C, Tzouvelekas V. Information transmission in irrigation technology adoption and diffusion. American Journal of Agricultural Economics. 2014;96(1):233-250.
12. Wossen T, Abdoulaye T, Alene A, Haile MG, Feleke S. Impacts of extension access and cooperative membership on technology adoption and household welfare. Journal of Rural Studies. 2017;54:223-233.
13. Valente TW, Davis RL. Accelerating the diffusion of innovations using opinion leaders. Annals of the American Academy of Political and Social Science. 1999;566(1):55-67.
14. Meinzen-Dick R, Quisumbing AR, Behrman J, Biermayr-Jenzano P, Wilde V. Gender and agriculture: closing the knowledge gap. Dordrecht: Springer Netherlands; 2014.
15. Katungi E, Edmeades S, Smale M. Gender, social capital and information exchange in rural Uganda. Journal of International Development. 2008;20(1):3552.
16. Conley TG, Udry CR. Learning about a new technology: pineapple in Ghana. American Economic Review. 2010;100(1):35-69.
17. Munshi K. Social learning in a heterogeneous population: technology diffusion in the Indian Green Revolution. Journal of Development Economics. 2004;73(1):185-213.
18. Abebaw D, Haile MG. The impact of cooperatives on agricultural technology adoption: empirical evidence from Ethiopia. Food Policy. 2013;38:82-91.
19. Hogset H, Barrett CB. Social learning, social influence, and projection bias: a caution on inferences based on proxy reporting of peer behavior. Economic Development and Cultural Change. 2010;58(3):563589.