This thesis investigates the relationship between organizational configurations and employee retention in new entrepreneurial ventures. Using a unique dataset of Italian firms in the transition-to-scale-up phase, it examines whether distinct configurations of structural, managerial and human-resource practices correspond to different turnover outcomes. A k-medoids cluster analysis, based on sixteen organizational variables, identifies six configurations: Organic, Hybrid Organic, CEO-Centric, Hybrid CEO-Centric, Quasi-Professionalized and Professionalized. Descriptive evidence shows that turnover levels are generally low, yet differences emerge across configurations. Ventures in the Quasi-Professionalized configuration display higher turnover frequency and intensity, while the Hybrid Organic and Hybrid CEO-Centric clusters exhibit the lowest turnover. The Professionalized configuration presents moderate and stable retention outcomes. However, when these descriptive patterns are examined through regression models, cluster membership does not display statistically significant explanatory power once sectoral and firm-level characteristics are controlled for. Sectoral affiliation, particularly in business services and in trade and transportation, emerges as the strongest and most consistent predictor of both the likelihood and intensity of turnover. Firm size and firm age show moderate, marginal effects on turnover probability in the logit model, larger firms being somewhat more likely to experience turnover, and older firms slightly less so, but these effects do not persist in the OLS model. Regional variables and all other firm-level controls exhibit no robust association with turnover. Taken together, these findings suggest that differences in retention across organizational configurations are visible descriptively, but not statistically significant once broader contextual and firm-level factors are considered. The results support the idea that structural and HR choices matter for retention, yet their measurable impact may be overshadowed by sector-specific labour-market dynamics. The study contributes to the literature on new ventures by integrating a configurational perspective with empirical evidence from an Italian institutional context and highlights the need for richer, longitudinal data to uncover causal mechanisms.
Questa tesi indaga la relazione tra le configurazioni organizzative e la retention dei dipendenti nelle new ventures. Utilizzando un dataset di imprese italiane nella prima fase di crescita, analizza se differenti combinazioni di caratteristiche strutturali, manageriali e di gestione delle risorse umane corrispondano a diversi esiti di turnover. Una cluster analysis k-medoids, basata su sedici variabili organizzative, identifica sei configurazioni: Organic, Hybrid Organic, CEO-Centric, Hybrid CEO-Centric, Quasi-Professionalized e Professionalized. Le evidenze descrittive mostrano che i livelli di turnover sono generalmente bassi, ma emergono differenze tra le configurazioni. Le imprese appartenenti alla configurazione Quasi-Professionalized presentano una maggiore frequenza e intensità di turnover, mentre le configurazioni Hybrid Organic e Hybrid CEO-Centric registrano i livelli più bassi. La configurazione Professionalized mostra invece una retention moderata e stabile. Tuttavia, quando questi pattern descrittivi vengono analizzati tramite modelli di regressione, l’appartenenza al cluster non risulta statisticamente significativa una volta controllati i fattori settoriali e le caratteristiche aziendali. L’appartenenza ad un settore, soprattutto nei servizi alle imprese e nel commercio e trasporti, emerge come il predittore più forte e coerente sia della probabilità sia dell’intensità del turnover. La dimensione dell’impresa e la sua età mostrano effetti moderati e marginali nel modello logit: le imprese più grandi risultano leggermente più propense a sperimentare turnover, mentre quelle più mature mostrano una probabilità leggermente inferiore; tali effetti non permangono però nel modello OLS. Le variabili regionali e gli altri controlli a livello d’impresa non evidenziano associazioni robuste. Nel complesso, questi risultati suggeriscono che, sebbene le configurazioni organizzative presentino differenze descrittive nei livelli di turnover, tali differenze non risultano statisticamente significative quando si considerano i fattori contestuali e aziendali più ampi. I risultati confermano che le scelte strutturali e di gestione delle risorse umane influenzano la retention, ma indicano anche che il loro impatto misurabile può essere attenuato dalle dinamiche settoriali del mercato del lavoro italiano. Questa tesi contribuisce alla letteratura esistente sulle new ventures integrando una prospettiva configurazionale con evidenze empiriche e sottolinea la necessità di dati longitudinali più ricchi per chiarire i meccanismi causali sottostanti.
Organizational configurations and employee retention in new ventures
SORLINI, CHIARA;SIMONCELLI, GRETA
2024/2025
Abstract
This thesis investigates the relationship between organizational configurations and employee retention in new entrepreneurial ventures. Using a unique dataset of Italian firms in the transition-to-scale-up phase, it examines whether distinct configurations of structural, managerial and human-resource practices correspond to different turnover outcomes. A k-medoids cluster analysis, based on sixteen organizational variables, identifies six configurations: Organic, Hybrid Organic, CEO-Centric, Hybrid CEO-Centric, Quasi-Professionalized and Professionalized. Descriptive evidence shows that turnover levels are generally low, yet differences emerge across configurations. Ventures in the Quasi-Professionalized configuration display higher turnover frequency and intensity, while the Hybrid Organic and Hybrid CEO-Centric clusters exhibit the lowest turnover. The Professionalized configuration presents moderate and stable retention outcomes. However, when these descriptive patterns are examined through regression models, cluster membership does not display statistically significant explanatory power once sectoral and firm-level characteristics are controlled for. Sectoral affiliation, particularly in business services and in trade and transportation, emerges as the strongest and most consistent predictor of both the likelihood and intensity of turnover. Firm size and firm age show moderate, marginal effects on turnover probability in the logit model, larger firms being somewhat more likely to experience turnover, and older firms slightly less so, but these effects do not persist in the OLS model. Regional variables and all other firm-level controls exhibit no robust association with turnover. Taken together, these findings suggest that differences in retention across organizational configurations are visible descriptively, but not statistically significant once broader contextual and firm-level factors are considered. The results support the idea that structural and HR choices matter for retention, yet their measurable impact may be overshadowed by sector-specific labour-market dynamics. The study contributes to the literature on new ventures by integrating a configurational perspective with empirical evidence from an Italian institutional context and highlights the need for richer, longitudinal data to uncover causal mechanisms.| File | Dimensione | Formato | |
|---|---|---|---|
|
2025_12_Simoncelli_Sorlini_Tesi.pdf
non accessibile
Descrizione: Testo della tesi
Dimensione
1.69 MB
Formato
Adobe PDF
|
1.69 MB | Adobe PDF | Visualizza/Apri |
|
2025_12_Simoncelli_Sorlini_Executive_Summary.pdf
non accessibile
Descrizione: Executive summary
Dimensione
777.4 kB
Formato
Adobe PDF
|
777.4 kB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/246097