Categories
Uncategorized

Research and Statistical Consulting

It has been now 5 years working in the area of statistical and research consulting and there are valuable lessons that have been learnt from numerous projects executed and the interactions with clients.

My expertise has been honed more as a quantitative researcher but over the years, one finds that qualitative skills of listening on the ground are as necessary than the carefully constructed answer options given as yes no answers or those given on Likert scales.

More than anything else, statistical and research consulting is all about understanding the client requirements and their limitations. Building that understanding requires having multiple conversations, and the numerous back and forth to arrive at alternative scenarios and answering tons of questions, why this is necessary and why this is not. The client is often uncertain, has limitations of bandwidth, is cash strapped and has to satisfy and pacify a number of internal and external constituencies.

In order to be able to satisfy the client, a research consultant must have the following qualities:

  1. Would like to solve real problems and also empower others to solve problems as well. This often requires the researcher to don the hat of capacity building as well
  2. Be an active listener and ask the required but direct questions in a way that enables the client to think deeply
  3. Be able to put forward the problem in clear and simple terms. This may often require multiple rounds of presentations
  4. Must have a broad knowledge and clear understanding of statistical and scientific methods
  5. Be willing to read, research existing literature and approaches that relate to the clients requirements and area of problem
  6. Be willing to apply existing statistical approaches and design to different environments
  7. Be able to write and communicate effectively the essence of the findings in a manner that the client is able to understand
  8. Ensure that estimates are statistically robust and that recommendations have taken a view of literature on that subject.
  9. Be prepared to give solutions that are actionable for the client while solving the clients’s problems in as much entirety as possible. This is something that needs an emphasis with a double underline.
  10. Avoid recommending to the client research shortcuts that compromise the quality of data or the quality of findings and will ultimately make the client look bad. Make sure that due process of research is followed with documentation of references and discussions
  11. Ensure that if possible get some experience in the actual collection of data or at least if you get only the data, ask questions of who collected the data, whether training was done, what steps were taken to avoid investigator bias, what sampling method was followed and so on
  12. Always take a bit of extra time to check and double check whether you have checked your results and procedures. Sometimes mistakes happen when you have been doing this over and over again and in the hurry to meet the client brief, you forget to double check that tiny coding result or the tiny analysis step. Remember, even it does not make a material difference to the result, still try and ensure that research standards are met.
  13. Be able to communicate effectively in writing as well as orally the final results. Often the client may be looking at the solution in a different way and you have to explain why your approach may be more reasonable and accurate. At the same time be flexible that the client has a point because at the end of the day, he is the one who is probably living and breathing the problem
  14. Be able to make a good estimate of how much effort will be required to solve the problem without actually having to solve the problem itself. That is sometimes the most difficult aspect because you have to bill the client. The client may not realize the intricacies of what is involved and the need to clean data, check its accuracy, run background checks, see what methodology works, performs statistical testing and robustness of estimates, write down results in a way that is accurate and understandable to the client. The client wants results fast, at rock-bottom prices and thinks that all of you have to probably wave a magic wand to get this done in a jiffy. That is the subject of persuasion, conversation and negotiation.

Research consulting is like any other consulting, but sometimes more arduous, because it is evidence backed and requires a fair degree of programming, testing and alternative ways of a looking, structuring and solving a problem. But the road map defined above may well help you to be a good research companion to your clients.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s