Data Authenticity and Integrity is the most important aspect of any survey. A whole lot of analysis in the data depends upon the set integrity achieved during data collection. To ensure that validation serves as an important tool for the survey.
Data Authenticity and Integrity is the most important aspect of any survey. A whole lot of analysis in the data depends upon the set integrity achieved during data collection. To ensure that validation serves as an important tool for the survey.
Validation - Validating a survey refers to the process of checking the survey responses for their validity. Because there are multiple, complex factors that can influence the dependability of a question. Validating a survey is quite an intricate task that needs our attention for getting the best results.
Types of validation are:
1. Force Response: It requires the respondent to answer the question before they can proceed to the survey.
2. Content Validation: It is used to check a certain response based on a certain type of content (such as when you want a respondent to enter a valid email address).
3. Custom Validation: It is used to check a certain type of response (for example, making sure a respondent selects the right amount of answer choices in a multi-select question).
Newer types of Validation:
Range-based Validation and Arithmetic Validation.
It is seen that range based validation is an important concept. To understand this we need to go through an example.
Supposing a farmer is cultivating crops which we may not know. A survey is to be undertaken to understand the livelihood conditions of the farmers. The requisite sample has been derived and the survey was undertaken keeping in mind certain parameters.
The parameters are:
1. Type of crops grown-It maybe rice, wheat, dal, etc.
2. Amount of farming land
3. Select Crops
4. Total production
5. The income of the farmer yearly basis.
The factors such as quantity, area, income are depended on the arithmetic calculations and range based validation.
The most important factor like Total production requires a substantiate range of numbers to be entered otherwise while doing the survey there can be numerous wrong number feeds which might hamper the result. So range based validation is assigned to check such wrongdoings and the correct entry is ascertained.
For example, on average, it is seen that Paddy production 25 Qty/acre and Wheat production 20 Qty/acre. Now, if someone says he has 3 acres of land and he produces Paddy, the accepted limit of total production is around 75 Qty with a +-10 percent accuracy. i.e. it can accept 70 Qty but not accept 50 Qty\
While calculating the yearly income of a farmer or the percentage of growth on a Y-o-Y basis, it requires three parameters such as the quantity of crop produced, the selling price of the crop in each month, and then converting into yearly yearly income. This can prove easy while calculating for one farmer but it can prove cumbersome while calculating for 100 odd farmers. So at that moment, the accurate data can only be received through strong validation in accepting the range of numbers or the length of numbers that is to be fed in the data collection system.
With the digitization of data, validation has become more useful and its use has become well known. So in order to continue validating a survey has become a matter of primary importance.