This thesis examines and extends the methodological foundations of Experience Sampling Method (ESM) research by developing systematic measurements, applying advanced analytical techniques, and exploring innovative solutions to distributional challenges. ESM enables real-time data collection in daily life, but presents unique challenges involving complex data structures, measurement development needs, and data characteristics requiring specialized approaches for accurate parameter estimation and valid conclusions. The thesis is divided into three parts: Part I describes systematic bottom-up development of Honesty-Humility measures across trait, motivation, and state levels; Part II explores interpersonal emotion dynamics in romantic couples using advanced dyadic ESM approaches; Part III addresses a fundamental methodological issue of skewed distributions and bounded psychological constructs in ESM studies. Chapters 2-4 form an integrated sequence developing and validating Honesty-Humility measures for cross-sectional and intensive longitudinal research, demonstrating a systematic, bottom-up approach that minimizes researcher bias while ensuring ecological validity and psychometric rigor. Chapter 2 presents the Adjective Checklist of Honesty (ACH), a 22-item measure capturing the facets of Honesty-Humility. Developed using extensive Italian adjectives refined through expert ratings, this tool demonstrated good factorial validity and meaningful associations with relevant constructs. Chapter 3 develops the Goals for Honesty/Dishonesty (GH/GD) questionnaire through participant-generated goal elicitation from over 9,000 textual responses, classified using natural language processing. The 78-item questionnaire revealed GH and GD as distinct, moderately negatively correlated motivational orientations. Critically, GD predicted actual dishonest behavior beyond traditional trait measures. Chapter 4 adapts GH/GD for intensive longitudinal research. In a 15-day ESM study with 5 prompts per day (N=198), multilevel confirmatory factor analyses confirmed the two-factor structure at within- and between-person levels. Momentary goal importance robustly predicted Honesty-Humility states contemporaneously and temporally. Latent growth curve modeling demonstrated that goal changes predicted trait-level change through state changes, evidencing bottom-up personality development. Chapter 5 examines emotional interdependence in romantic relationships through an extended Longitudinal Actor-Partner Interdependence Model framework. Using data from 76 couples who completed assessments five times daily over 30 days, the results revealed significant positive affect interdependence both contemporaneously and temporally, but no negative affect interdependence. Contact enhanced males' positive affect but didn't moderate interdependence strength. Substantial between-couple variability highlighted that emotional interdependence is more complex and heterogeneous than theoretical models assume. Chapter 6 addresses skewed ESM data exhibiting floor/ceiling effects. Conceptualizing skewed observations as censored manifestations of latent variables, mixed-effects tobit models demonstrated superior parameter recovery versus standard linear models. Applied to artificially censored positive affect and naturally skewed negative affect (N=322), tobit models estimated larger effect sizes, greater between-person variability, and more plausible parameter relationships than standard approaches. The thesis demonstrates that assessing psychological variability through intensive longitudinal methods is crucial for understanding daily life dynamics. The systematic procedures, advanced techniques, and innovative solutions provide foundations for more accurate, nuanced, and replicable ESM research. Future developments should focus on multimodal integration, personalized study design, and continued analytical refinement for complex psychological phenomena.
Questa tesi esamina ed estende le basi metodologiche della ricerca basata sull’Experience Sampling Method (ESM), sviluppando misurazioni sistematiche, applicando tecniche analitiche avanzate e proponendo soluzioni innovative alle sfide legate alle distribuzioni dei dati. L’ESM consente la raccolta di dati in tempo reale nella vita quotidiana, ma presenta difficoltà legate a strutture di dati complesse, sviluppo di misure adeguate e caratteristiche che richiedono approcci specializzati per stime accurate e conclusioni valide. La tesi è articolata in tre parti. La Parte I descrive lo sviluppo sistematico “bottom-up” di misure di Onestà-Umiltà a livello di tratto, motivazione e stato. La Parte II esplora le dinamiche emotive interpersonali nelle coppie romantiche tramite approcci ESM diadici avanzati. La Parte III affronta un problema metodologico fondamentale: le distribuzioni asimmetriche e i costrutti psicologici "bounded" negli studi ESM. I Capitoli 2-4 sviluppano e validano strumenti di Onestà-Umiltà per ricerche trasversali e longitudinali intensive, mostrando un approccio sistematico che riduce i bias e garantisce validità ecologica e rigore psicometrico. Il Capitolo 2 presenta l’Adjective Checklist of Honesty (ACH), una misura di 22 item basata su aggettivi italiani selezionati da esperti, con buona validità fattoriale e correlazioni significative con costrutti affini. Il Capitolo 3 introduce il questionario Goals for Honesty/Dishonesty (GH/GD), sviluppato da oltre 9.000 risposte testuali analizzate con tecniche di linguaggio naturale. Le due dimensioni, GH e GD, risultano distinte e moderatamente correlate negativamente; in particolare, GD predice comportamenti disonesti oltre le misure di tratto tradizionali. Il Capitolo 4 adatta GH/GD alla ricerca longitudinale intensiva. In uno studio ESM di 15 giorni con 5 rilevazioni al giorno (N=198), analisi multilivello hanno confermato la struttura bifattoriale e mostrato che l’importanza momentanea degli obiettivi predice gli stati di Onestà-Umiltà nel tempo. Latent Growth Model indica che i cambiamenti negli obiettivi producono variazioni di tratto attraverso i cambiamenti di stato, evidenziando uno sviluppo “bottom-up” della personalità. Il Capitolo 5 indaga l’interdipendenza emotiva nelle relazioni romantiche mediante un’estensione del Longitudinal Actor-Partner Interdependence Model (L-APIM). Su dati di 76 coppie valutate cinque volte al giorno per 30 giorni, emerge una significativa interdipendenza dell’affetto positivo, ma non di quello negativo. Il contatto aumenta l’affetto positivo maschile ma non modifica la forza dell’interdipendenza. La variabilità tra coppie indica che i legami emotivi sono più complessi di quanto previsto dai modelli teorici. Il Capitolo 6 affronta le distribuzioni asimmetriche tipiche dei dati ESM, con effetti di pavimento o soffitto. Considerando le osservazioni come manifestazioni censurate di variabili latenti, i modelli mixed-effects tobit mostrano un recupero dei parametri superiore rispetto ai modelli lineari. Applicati a dati di affetto positivo censurati artificialmente e di affetto negativo naturalmente asimmetrici (N=322), stimano effetti più realistici e maggiore variabilità interindividuale. La tesi dimostra che valutare la variabilità psicologica con metodi longitudinali intensivi è essenziale per comprendere le dinamiche quotidiane. Le procedure sistematiche, le tecniche avanzate e le soluzioni innovative qui proposte gettano le basi per una ricerca ESM più accurata, sfumata e replicabile. Gli sviluppi futuri dovrebbero concentrarsi sull'integrazione multimodale, sulla progettazione personalizzata degli studi e sull'ulteriore perfezionamento analitico per affrontare fenomeni psicologici complessi.
Galkina, A (2026). Dear Diary: Advances in Experience Sampling Methodology Studies. (Tesi di dottorato, , 2026).
Dear Diary: Advances in Experience Sampling Methodology Studies
GALKINA, ANASTASIIA
2026
Abstract
This thesis examines and extends the methodological foundations of Experience Sampling Method (ESM) research by developing systematic measurements, applying advanced analytical techniques, and exploring innovative solutions to distributional challenges. ESM enables real-time data collection in daily life, but presents unique challenges involving complex data structures, measurement development needs, and data characteristics requiring specialized approaches for accurate parameter estimation and valid conclusions. The thesis is divided into three parts: Part I describes systematic bottom-up development of Honesty-Humility measures across trait, motivation, and state levels; Part II explores interpersonal emotion dynamics in romantic couples using advanced dyadic ESM approaches; Part III addresses a fundamental methodological issue of skewed distributions and bounded psychological constructs in ESM studies. Chapters 2-4 form an integrated sequence developing and validating Honesty-Humility measures for cross-sectional and intensive longitudinal research, demonstrating a systematic, bottom-up approach that minimizes researcher bias while ensuring ecological validity and psychometric rigor. Chapter 2 presents the Adjective Checklist of Honesty (ACH), a 22-item measure capturing the facets of Honesty-Humility. Developed using extensive Italian adjectives refined through expert ratings, this tool demonstrated good factorial validity and meaningful associations with relevant constructs. Chapter 3 develops the Goals for Honesty/Dishonesty (GH/GD) questionnaire through participant-generated goal elicitation from over 9,000 textual responses, classified using natural language processing. The 78-item questionnaire revealed GH and GD as distinct, moderately negatively correlated motivational orientations. Critically, GD predicted actual dishonest behavior beyond traditional trait measures. Chapter 4 adapts GH/GD for intensive longitudinal research. In a 15-day ESM study with 5 prompts per day (N=198), multilevel confirmatory factor analyses confirmed the two-factor structure at within- and between-person levels. Momentary goal importance robustly predicted Honesty-Humility states contemporaneously and temporally. Latent growth curve modeling demonstrated that goal changes predicted trait-level change through state changes, evidencing bottom-up personality development. Chapter 5 examines emotional interdependence in romantic relationships through an extended Longitudinal Actor-Partner Interdependence Model framework. Using data from 76 couples who completed assessments five times daily over 30 days, the results revealed significant positive affect interdependence both contemporaneously and temporally, but no negative affect interdependence. Contact enhanced males' positive affect but didn't moderate interdependence strength. Substantial between-couple variability highlighted that emotional interdependence is more complex and heterogeneous than theoretical models assume. Chapter 6 addresses skewed ESM data exhibiting floor/ceiling effects. Conceptualizing skewed observations as censored manifestations of latent variables, mixed-effects tobit models demonstrated superior parameter recovery versus standard linear models. Applied to artificially censored positive affect and naturally skewed negative affect (N=322), tobit models estimated larger effect sizes, greater between-person variability, and more plausible parameter relationships than standard approaches. The thesis demonstrates that assessing psychological variability through intensive longitudinal methods is crucial for understanding daily life dynamics. The systematic procedures, advanced techniques, and innovative solutions provide foundations for more accurate, nuanced, and replicable ESM research. Future developments should focus on multimodal integration, personalized study design, and continued analytical refinement for complex psychological phenomena.| File | Dimensione | Formato | |
|---|---|---|---|
|
phd_unimib_851053.pdf
accesso aperto
Descrizione: Dear Diary: Advances in Experience Sampling Methodology Studies
Tipologia di allegato:
Doctoral thesis
Dimensione
5.34 MB
Formato
Adobe PDF
|
5.34 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


