I have 78 variables in it. Need to rank and prioritize 78 variables based on the opinion of experts.
Initially thought of starting with Fuzzy AHP, DEMETEL etc but found that FAHP is good for small number of variables.
Then I thought of to go for EFA (factor analysis).
My question is:
1. Which technique can be used to rank and prioritize these 78 variables?
2. Do we have any alternative or big data version of Fuzzy AHP to do the analysis?
3. If yes, how to draft a questionnaire for such 78 variables, their interrelationship with each other?
Kindly share your opinion.
Dear Manvendra pratap Singh,
Based on the information provided, here are some suggestions for the given problems:
1. Ranking and prioritizing 78 variables:
To rank and prioritize the 78 variables, you can consider using the Analytic Hierarchy Process (AHP). AHP is a widely used technique for multi-criteria decision-making that allows for the systematic comparison and prioritization of different factors or variables. It involves constructing a hierarchical structure, establishing pairwise comparisons between variables, and calculating weights based on the judgments. AHP provides a consistent framework for decision-making and allows for the synthesis of subjective opinions.
2. Alternative or big data version of Fuzzy AHP:
Fuzzy AHP is a modification of the traditional AHP that incorporates fuzzy logic to handle vagueness and uncertainty in decision-making. While there may not be a specific big data version of Fuzzy AHP, you can explore using big data analytics techniques, such as machine learning algorithms or data mining approaches, to analyze large datasets and extract insights related to ranking and prioritization. These techniques can handle massive amounts of data and uncover patterns or relationships that may not be easily identifiable through traditional statistical methods.
3. Drafting a questionnaire for the variables' interrelationship:
When drafting a questionnaire to capture the interrelationships between the 78 variables, you can consider the following steps:
a. Determine the purpose: Clearly define the objective of the questionnaire and what information you aim to gather regarding the interrelationships between the variables.
b. Identify key variables: Identify the key variables that you want to explore in terms of their interrelationships. You may consider conducting a preliminary analysis or literature review to identify the most relevant variables.
c. Establish the structure: Structure the questionnaire in a way that allows participants to provide their assessments of the interrelationships. This can be done through rating scales, ranking exercises, or pairwise comparisons.
d. Develop questions: Create specific questions that ask participants to assess the strength or direction of the interrelationships between the variables. Ensure that the questions are clear, concise, and unbiased.
e. Pilot testing: Before administering the questionnaire to the target participants, conduct a pilot test with a small sample to identify any issues with the clarity or effectiveness of the questions. Revise and refine the questionnaire based on the feedback received.
f. Administer the questionnaire: Administer the questionnaire to the intended participants, ensuring that instructions are provided clearly and participants have enough time to complete the questionnaire.
g. Analyze the responses: Once you have collected the responses, analyze the data using appropriate statistical techniques, such as correlation analysis or fuzzy logic approaches, to assess the interrelationships between the variables.
Remember to consider ethical considerations, maintain confidentiality, and obtain informed consent from participants when conducting any data collection activities.
These suggestions should provide you with a starting point for addressing the given problems. However, it is important to note that the specific details and requirements of your research may influence the choice of techniques and approaches.
If you want to develop a questionnaire, first cycle back to your research question. One questionnaire with 78 variables would be a lengthy document which would necessitate a large sample (my point is one questionnaire for so many variables is uncommon and probably should be multiple scales). I suggest you start with the singular purpose (e.g., grit has two factors, but the major objective is grit, etc.) and then do the following: 1.) Examine the extant literature for scales in your area already developed; 2.) Use theory to develop questions; 3.) Consult other experts, possibly a Delphi technique; 4.) Pilot and review. Then you might be ready for EFA/CFA.
One last point: The downside to AHP is one must assume all relationships are hierarchical. There could be familial, tangential, etc.
- Badges
- Science topic
- Similar topics
- Correction