Perhaps you might like what this paper does: Article Duality in Diversity: How Intrapersonal and Interpersonal Cu...

They use topic modeling in the study of (organizational) culture.

Corporate narrative research considers inter alia how an organization attempts to convince through argumentation, including what strategies it has to convince audiences; how speakers and writers endeavor to inform, persuade, or motivate particular audiences in particular situations; how arguments are justified or validated; what kinds of vocabularies, expressions, and metaphors are used; how speakers and writers position themselves and others; and how the positions of speakers and writers are authorized.

Moretti and Pestre (2015) conducted a quantitative linguistic analysis of more than 65 years of the World Bank’s annual reports and found a sharp decline in factual precision, replaced by what they termed "management discourse". Moretti and Pestre (2015) do not claim to measure organizational culture sensu stricto but the mapping of changes in the World Bank’s language, syntax, and grammar over time does provide evidence of cultural change of a kind found in other international organizations too.

Reference

Moretti, F. & Pestre, D. (2015). Bankspeak: The language of World Bank reports, 1946–2012. New Left Review 92. Retrieved from https://newleftreview.org/II/92/franco-moretti-dominique-pestre-bankspeak

We measure culture as part of a whole system research diagnostic that is explicitly linked with value outcomes (financial, environmental). Our question set requires qualitative and quantitative evidence and produces a numerical outcome using a credit rating type scale. More information here:

http://www.hrmaturity.com/cambridge-judge-business-school-and-mi-embark-on-ground-breaking-joint-research-project/

Whilst most approaches to this question can come up with buckets of data, often, the resultant data can be skewed, largely due to the respondent becoming a bit tentative due to the sensitive nature of the purpose (measuring organisational culture). What we have used, often as a validation of typical and customary measurements that usually come from a HR department, is a far simpler set of questions that represent an analogy. By using an analogy, the respondents feel more comfortable about saying what they really might believe about an organisational culture. The questions used with the analogy are:

1) If our organisational culture was some type of vehicle, what vehicle would it be? (Note that a "vehicle" is something made by humans that goes from point A to point B.)

2) What condition is the vehicle in? (here you want detail about the condition of the vehicle; i.e. colour, cleanliness, maintenance, dents or prangs, etc.)

3) What part of the vehicle are you? (a respondent cannot be a passenger or driver in a vehicle, but must be an integral part of the vehicle itself.)

4) In 5 years, what kind of vehicle do you expect it to be? (the 5 years could be 3 years or any other timeframe that could be important)

In many cases, respondents have identified an organisational culture as an auto, but equally, other vehicles that have been chosen are airplanes, buses, trains, and skateboards...the span of vehicles chosen is quite wide. Question 2 is all about getting past what the organisation would like others to perceive the culture is like. The answers to question 3 are quite interesting as they show how respondents feel about their positions within an organisational culture and structure. The last question illuminates the perceptions regarding whether or not the respondents feel that the organisational culture will grow positively over time or remain more or less the same as it is currently.

Clearly, these questions are best applied in a facilitated environment and like many things in life, do require a bit of practice and a facilitator who has a broad knowledge base of vehicles that might surface in a reply.

Thank you for the answers Richard Haans , Olivier Serrat , Stuart Woollard , James B. Rieley .