How not to present survey data- 2017 UN Global Staff Survey edition

Last week the UNOG Staff Coordinating Council shared the results of its 2017 Global Staff Satisfaction Survey 2017, “in which 4,000 of you took part last week, representing ten percent of staff.”

As a social science researcher I can confirm that this is a very sad case study of “how not to present survey results” and clearly not worthy of the UN system, its staff and the important issues that are raised in the report. Some of my concerns are simply about the poor presentation of findings (in the sense that most advanced undergrad students of pretty much any discipline should know better), but there are more concerning issues with the results that border on unethical (social) science research practices.

First of all, sharing an unedited 200 page pdf-document that does neither list the questions that were asked in the survey nor includes basic demographic data, e.g. distribution of respondents along organizations, gender, age, country, employment level etc. makes it difficult to assess the findings and put them into perspective.
The document contains a very brief outline of the methodology (p.6), but that still does not help in assessing how data was collected for 29 questions (probably too many based on my experience, especially if questions had sub-questions).
As there is no executive summary or any other interpretation of the results the reader is left with a lot of guess work.

Pages 37-204 of the document consist of additional comments that were added as free text. I have not had time to read all of them, let alone process them in any way, but that's not the point of this post anyway.
At the very least it is not best practice to simply share an unedited account of these responses, keeping in mind that they represent a very small fraction of the 4000 respondents.
Some of the comments are actually quite detailed, organization- or location-specific and some may consider it unethical to publish them-especially as respondents may not have been aware that their comments will show up verbatim in a document on the Internet.

“The answers may confirm your impressions but will more likely surprise you.”-nice Buzzfeed-style clickbaiting teaser-but there are actually some major problems in publishing un-analyzed data. Some of the comments range from general themes along the lines of “the UN is sooooo bureaucratic” to very specific issues around workplace safety in certain buildings to outright criminal accusations (“our organization is rife with corruption”) and general UN-bashing, for example on page 37:

Cronyism has become rampant in DESA, with appointments at D2 and D1 levels based on servility of supervisors, and very little merit. The resulting mediocre leadership is therefore incompetent, dearly lacking in vision, effectiveness and efficiency, is sometimes abusive, and has destroyed the morale of staff across several divisions.
I even found some names in the comments-luckily mostly linked to positive feedback so the person may not encounter negative repercussions at her/his workplace.
Sufficient to say this is all pretty bad.

All in all, I find the presentation and lack of
analysis of the findings simply un-worthy of the UN, its organizations and staff members who can clearly do a better job presenting the data and starting a meaningful discussion.
This is the first time that such a survey has been carried out at the UN in recent memory. As such, there will be room for improvement in the questions and methodology. lessons will be learned and much debate will be had.
Ok, that’s not enough of a disclaimer, I’m afraid. The results are publicly available including to people or journalists who are not as well-intentioned as I might be.
Without proper framing and analysis this survey can backfire as a rumor-mill or to spread anti-UN sentiments and I do not see how this will help to create a meaningful dialogue-let alone changes for staff members in difficult organizational contexts.

Popular posts from this blog

Reading #Harvey through a #globaldev lens

Links & Contents I Liked 247

Links & Contents I Liked 248

Links & Contents I Liked 235

Links & Contents I Liked 249