Cluster analysis with binary data

[Lena Oetting]

Dear statalist members,
in my current project I try to find a definition for Work 4.0.

I have the following variables
di

gital technology (Network = 1 data Analysis = 2 communication = 3 virtualization= 4)

key actor (human = 1 human-system = 2 system = 3)

worktask (manual routine = 1 manual non-routine = 2 cognitive routine = 3 cognitive non- routine = 4)

s

ituation (activity = 1 process = 2 structure = 3)

challenge

/potential (time = 1 costs = 2 quality = 3)

The variables and their characteristics are each allocated to a project within the scope of Work 4.0 (by ID; ID has for the time being only 12 observations). Now I try to Analyse which characteristics occur frequently together in the projects and form a pattern in order to provide a Definition for the structure of Work 4.0 projects.

So I would like to find out sth like:
"A Project can be considered as Work 4.0 if the characteristics Network, human-System, cognitive non-Routine, structure and quality are prevalent in the respective Project."

So the Question is: What is Work 4.0 and how can I derive a definition by means of currently existing project data?

I currently think of conducting a cluster analysis by dichotomizing the characteristics. Does this approach make sense for finding patterns in data or should I try a descriptive analysis?


I appreciate any suggestion.
Kind regards
Lena Oetting

[Phil Bromiley]

You didn't get a quick answer. You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex.

You can hit this with a variety of multivariate statistics tools, or you can try to do it manually. If you go the manual route, you should look at Tukey's work on exploratory data analysis.