I want to conduct a pilot study to ensure validity of adopted instrument in my context and parsimony of instrument. I need some guidance to what steps I should consider.
Hi Sadia
I suggest you sit down with a few (3-4) people, and ask them to complete the questionnaire with you next to them. You can then check with them for things like readability, flow of questions. The pilot sample should be typical of the likely target population.
thanks sir for your kind response sir i want to conduct a pilot study to ensure validity of instrument in my context
Dear Sadia Babar: First of all you need to develop a good survey instrument which addresses the issues you want to investigate, test it with a much smaller sample to see if it works and will generate the required data that you will need for your analysis. I am sure you will find several issues which you would like to address before you can take it as final. Take help of some experienced people in the process who can suggest ways to improve your survey instrument. Once it is final. you may go ahead with your pilot study data collection exercise. Once the data are collected, you will need to tabulate them and see what quantitative techniques can be applied and how, which would lead you to the final conclusion.
thanks sir for your detail response i have circulated my instrument for genrating response for my pilot study
In my previous study on the formulation supply chain integration strategy for global manufacturers based on a sample of 558 manufacturers in 17 countries, I have developed a comprehensive research process that covers both instrument development and model testing. Here are the details:
So, S. and Sun, H. (2010). Supplier integration strategy for lean manufacturing adoption in electronic-enabled supply chains. Supply Chain Management: An International Journal, 15(6), 474-487. It is available for download at this site:
https://www.researchgate.net/publication/235302947_Supplier_integration_strategy_for_lean_manufacturing_adoption_in_electronic-enabled_supply_chains?ev=prf_pub
Article Supplier integration strategy for lean manufacturing adoptio...
Depending on your sample, are there any 'experts' that you can discuss your instrument with? For example, in developing a dementia diary, I spoke to a group of carers who, although they had no formal qualifications, I believe were 'experts' in caring for a dementia patient who was a close family member. In discussion with them, I saw the need for additional fields that I had not considered, and fields that were not of any use. By speaking directly with your potential sample population, you can tease out what is best to focus on. Hope that helps.
Hi Sadia, you might need to consider the following if you are doing a social science research:
1) evaluate your prepared survey questionnaire with subject matter experts, scholars including your supervisor (if you are doing some post-graduate courses) so that they can highlight correction needed in your questionnaire e.g. grammar, multi-barreled questions, multi-lingual questions compatibility, questions clarify etc.
2) test your corrected questionnaire with some convenient sample you can obtained e.g. minimal 30 and above samples (suggest you sit beside them when they try to answer your questions so that you can clarify anything if they don't understand or misunderstood the questionnaire)
3) run some basic statistic tests e.g. Cronbach-Alpha Reliability, Composite Reliability etc. on the questionnaire instrument. Also based on your conceptual framework / model, run some inferential statistic tests according to your hypotheses stated e.g. multiple-regression, path analysis, t-test, ANOVA etc. to see what is the pilot result looks like.
4) after done the above no. 1-3, you can further revise your questionnaire and plan further how to source the right & adequate random samples etc for your actual data collection.
All the best.
Regards,
Fung
- Badges
- Science topic
- Similar topics
- Statistics
- Survey