This research explores how B Corp certified companies communicate sustainability within their official reports, with the aim of evaluating the consistency between the textual content and the scores obtained in the five areas of the B Impact Assessment (Governance, Workers, Community, Environment, Customers). The study takes place in a context of growing attention to the transparency and reliability of sustainability claims, particularly by organizations that are formally committed to creating value for society and the environment. The analysis was conducted on a corpus of ten sustainability reports published by companies operating in different sectors, all of which are B Corp certified. The research adopts a statistical text mining approach, combining computational linguistics tools with quantitative methods to systematically explore the language and communication practices used in corporate documents. Text data processing followed a structured workflow that included text cleaning and normalization, sentence segmentation, thematic labeling of information units based on the B Impact Assessment areas, and frequency analysis. The tools used included AntConc, IRaMuTeQ, RStudio, and, in the initial phase, Python, which was used exclusively for extracting text from PDF documents. Particular attention was given to identifying keywords, analyzing the distribution of themes, and detecting recurring linguistic patterns, with the aim of measuring the emphasis and centrality of the various dimensions of sustainability. In addition to the quantitative component, an in-depth qualitative analysis of selected sentences was conducted, with the aim of contextualizing the data and identifying possible communication strategies, omissions, or inconsistencies. The results of the textual analyses were then compared with the scores assigned to companies in the respective B Impact Assessment areas, in order to assess the degree of alignment between what is reported and the certified performance. The general objective of this study is to understand whether, and to what extent, the language used in sustainability reports faithfully reflects the actual practices and results of B Corp companies, or whether there are elements of discontinuity, selective narratives, or significant misalignments. The findings offer useful insights to encourage more transparent, consistent, and responsible communication, contributing to the academic and professional debate on sustainable reporting by companies engaged in concrete paths of social and environmental improvement.
Questa ricerca esplora in che modo le aziende certificate B Corp comunicano la sostenibilità all’interno dei loro report ufficiali, con l’obiettivo di valutare la coerenza tra i contenuti testuali e i punteggi ottenuti nelle cinque aree del B Impact Assessment (Governance, Lavoratori, Comunità, Ambiente, Clienti). L’indagine si inserisce in un contesto di crescente attenzione verso la trasparenza e l’affidabilità delle dichiarazioni di sostenibilità, specialmente da parte di organizzazioni che si impegnano formalmente a generare valore per la società e l’ambiente. L’analisi è stata condotta su un corpus composto da dieci report di sostenibilità pubblicati da imprese operanti in differenti settori, tutte accomunate dalla certificazione B Corp. Il lavoro si basa su un approccio di analisi statistica dei dati testuali, che combina strumenti di linguistica computazionale con metodi quantitativi, al fine di esplorare in modo sistematico il linguaggio e le pratiche comunicative adottate nei documenti aziendali. L’elaborazione dei dati testuali ha previsto una serie di fasi: pulizia e normalizzazione del testo, suddivisione in frasi, etichettatura tematica delle unità informative in base alle aree del B Impact Assessment, e analisi delle frequenze. Tra gli strumenti utilizzati figurano AntConc, IRaMuTeQ, RStudio e, in fase iniziale, Python, impiegato esclusivamente per l’estrazione dei testi dai documenti PDF. Particolare attenzione è stata rivolta all’identificazione delle parole chiave, all’analisi della distribuzione dei temi e alla rilevazione di pattern linguistici ricorrenti, con l’intento di misurare il grado di enfasi e centralità delle diverse dimensioni della sostenibilità. Accanto alla componente quantitativa, è stata svolta un’analisi qualitativa approfondita delle frasi selezionate, con l’obiettivo di contestualizzare i dati e identificare possibili strategie comunicative, omissioni o incongruenze. I risultati delle analisi testuali sono stati successivamente confrontati con i punteggi assegnati alle aziende nelle rispettive aree del B Impact Assessment, al fine di verificare il livello di allineamento tra ciò che viene dichiarato nei report e le performance certificate. L’obiettivo generale dell’indagine è comprendere se e in che misura il linguaggio utilizzato nei report di sostenibilità rifletta in modo fedele le pratiche e i risultati effettivi delle imprese B Corp, oppure se emergano elementi di discontinuità, narrazioni selettive o disallineamenti significativi. I risultati offrono spunti di riflessione utili per stimolare una comunicazione più trasparente, coerente e responsabile, contribuendo al dibattito scientifico e professionale sulla rendicontazione sostenibile da parte delle aziende impegnate in percorsi concreti di miglioramento sociale e ambientale.
Esplorazione dei report aziendali: analisi testuale comparativa della comunicazione della sostenibilità nelle aziende certificate B Corp
CARUSO, MARIA BERNADETTE
2024/2025
Abstract
This research explores how B Corp certified companies communicate sustainability within their official reports, with the aim of evaluating the consistency between the textual content and the scores obtained in the five areas of the B Impact Assessment (Governance, Workers, Community, Environment, Customers). The study takes place in a context of growing attention to the transparency and reliability of sustainability claims, particularly by organizations that are formally committed to creating value for society and the environment. The analysis was conducted on a corpus of ten sustainability reports published by companies operating in different sectors, all of which are B Corp certified. The research adopts a statistical text mining approach, combining computational linguistics tools with quantitative methods to systematically explore the language and communication practices used in corporate documents. Text data processing followed a structured workflow that included text cleaning and normalization, sentence segmentation, thematic labeling of information units based on the B Impact Assessment areas, and frequency analysis. The tools used included AntConc, IRaMuTeQ, RStudio, and, in the initial phase, Python, which was used exclusively for extracting text from PDF documents. Particular attention was given to identifying keywords, analyzing the distribution of themes, and detecting recurring linguistic patterns, with the aim of measuring the emphasis and centrality of the various dimensions of sustainability. In addition to the quantitative component, an in-depth qualitative analysis of selected sentences was conducted, with the aim of contextualizing the data and identifying possible communication strategies, omissions, or inconsistencies. The results of the textual analyses were then compared with the scores assigned to companies in the respective B Impact Assessment areas, in order to assess the degree of alignment between what is reported and the certified performance. The general objective of this study is to understand whether, and to what extent, the language used in sustainability reports faithfully reflects the actual practices and results of B Corp companies, or whether there are elements of discontinuity, selective narratives, or significant misalignments. The findings offer useful insights to encourage more transparent, consistent, and responsible communication, contributing to the academic and professional debate on sustainable reporting by companies engaged in concrete paths of social and environmental improvement.| File | Dimensione | Formato | |
|---|---|---|---|
|
TESI Definitiva - Maria Bernadette Caruso.pdf
Accesso riservato
Dimensione
5.15 MB
Formato
Adobe PDF
|
5.15 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/88313