The data for different indicators of the index was received in varied scales with different units and dimensions of measurement. Therefore, to make it comparable, the raw data was to be converted into values of 0 to 1 using a systemic and structured methodology-normalisation.
For this purpose, the Dimensional Index Methodology was used to normalise variables and attain scores for Districts based on their performance along 68 indicators, which were then compiled into 10 sectors.
The Dimensional Index Methodology is the most commonly used process for normalisation of values and subsequent ranking. In this method, the normalised value for each indicator is obtained by subtracting the minimum value among set from the raw value of indicators and then dividing it the data range (maximum – minimum value). The maximum and minimum values for each indicator are determined on the basis of raw values for that indicator across districts. This approach was specifically adopted to allow comparison across all 75 districts and generate overall ranks.
The following two equations have been used to normalise the indicator values:
| For Positive indicator Score = | x-min |
|---|---|
| max-min |
| For Negative indicator Score = | max-x |
|---|---|
| max-min |
Where:
Positive Indicator = for which Higher value is better
Negative Indicator = for which Lower value is better
Max = Highest indicator value among the districts
Min = Lowest indicator value among the districts
After the normalisation of raw data for all indicators the normalised value for each of them was multiplied with their respective weightage to obtain the indicator value.