Hi Monisha
Thanks for your questions. As well as reading the article that Becky has recommended, I would ask:

- Why are you planning to survey field staff (as distinct from interview them)? 
- Why are you planning to interview service users (as distinct from surveying them)?

The selection of methods should always relate to the purpose for which you are collecting the data, the analysis you hope to undertake, and the practical issues affecting your study (how many respondents there will be in each group, literacy levels, language issues, respondent burden, time, cost, travel, translation, etc etc.  If 'survey' implies quantitative data, will you have enough respondents to subject your data to the quantitative tests you want to use? Do you really already know enough to move from qualitative to quantitative data collection? These same kinds of questions also inform the decision about whether to use open or closed questions). 

Should you use the same questions with staff as with managers? There's no way to be sure without knowing what you asked managers, but I suspect not. The realist principle here is: Different people know about different aspects of your theory, because of the roles they hold. You should therefore ask staff about the bits they can tell you about, managers about the bits they know about, and service users about the bits they know about.  

But also, coming back to purpose: when you interviewed managers, you were building theory. Next you'll be testing it. Different purposes probably imply different questions.

There's lots of very good material on line about how to develop surveys in general, and most of that will apply to realist surveys as well.  It might be useful to use that to get a general sense of whether and how to go about it. Then mentally check each aspect of the 'advice for how to do it' against two things - realist principles; and the purpose for which you are collecting this specific piece of information. If it's consistent with all three, go for it. If not, something needs redesigning (and the only bit that can't be redesigned is the realist principles!)

Best of luck
Gill 


-----Original Message-----
From: Monisha Vaid [mailto:[log in to unmask]] 
Sent: Tuesday, 19 September 2017 7:40 PM
To: [log in to unmask]; Gillian Westhorp <[log in to unmask]>
Cc: Monisha Vaid <[log in to unmask]>
Subject: Re: realist review quesion

Hi Gill

Thank you for recommending this platform! 
I am also a PhD student at LaTrobe University using Realist Evaluation to understand Outreach Model of family planning service delivery in India. I have a couple of questions re the method of data collection. I am planning to use mixed method. I am developing an interview guide for the management staff to understand the theory of change highlighting context and mechanism that results in expected outcomes. In phase 2 of testing the program theories, I plan to conduct a survey with the field staff and interviews with service users (women who received contraceptives from outreach). But I am not sure how to develop a questionnaire for the field staff? Should the questionnaire have the same questions as asked from the managers during the theory development phase? Should it be close ended with options taken from the responses of the managers? 
Is there any reference tools (questionnaires) that I can refer to? I did refer the RAMASES interview guide to develop my program theory interview guide.

Please guide!

Regards
Monisha