Normalize the scores for metrics
You can use the Maximum normalization input setting to use normalized values to calculate assessment scores for questions (metrics).
When Maximum normalization input applies
The Maximum normalization input field appears only when:
- The data type of the question is either Choice or Multiple Selection.
- The Scored check box is not selected.
To use normalized scoring, the value assigned to each metric definition (possible answer) for each metric (question) must be unique.
In the following examples, the Scale definition is High (larger numerical values are good).
Choice type questions
- When Maximum normalization input is not selected
Image omitted: tprm-q-choice-no-max-norm-input.png
Option to select maximum normalization input.
Option to select maximum normalization input.
Formula: \(\[Value of the response\] – \[Lowest value\] \) / \( \[Highest value\] – \[Lowest value\]\) \* 100.
In this example, the question allows a choice among answers with values of `1`, `2`, and `4`.
|Response|Calculation|Score|
|--------|-----------|-----|
|Dog|\[\(1-1\)/\(4-1\)\] \* 100|0|
|Cat|\[\(2-1\)/\(4-1\)\] \* 100|33|
|Goldfish|\[\(4-1\)/\(4-1\)\] \* 100|100|
- When Maximum normalization input is selected
Image omitted: tprm-q-choice-max-norm-input.png
Option to select maximum normalization input.
Option to select maximum normalization input.
Formula: `([Value of the response]/ [Highest value]) * 100`.
In this example, the question allows a choice among answers with values of `1`, `2`, and `4` and normalization input values of `3`, `5`, and `9` respectively.
|Response|Calculation|Score|
|--------|-----------|-----|
|Dog|\(1 / 4\) \* 100|25|
|Cat|\(2 / 4\) \* 100|50|
|Goldfish|\(4 / 4\) \* 100|100|
Multiple selection type questions
- When Maximum normalization input is not selected
Image omitted: tprm-q-multi-sel-no-max-norm-input.png
Option to select maximum normalization input.
Option to select maximum normalization input.
Formula: `([Sum of the normalization input values for the selected responses] / [Sum of all the normalization input values]) * 100`.
In this example, the question allows for multiple selections among answers with values of `1`, `2`, and `4` and normalization input values of `3`, `5`, and `9` respectively.
|Response|Calculation|Score|
|--------|-----------|-----|
|Dog and Cat|\(\[3+5\] / 17\) \* 100|47|
|Cat|\(5 / 17\) \* 100|29|
|Cat and Goldfish|\(\[5+9\] / 17\) \* 100|82|
|Goldfish|\(9 / 17\) \* 100|53|
- When Maximum normalization input is selected
Image omitted: tprm-q-multi-sel-max-norm-input.png
Option to select maximum normalization input.
Option to select maximum normalization input.
- The system uses the value `0` for the response that has the lowest value. In this example, the **Dog** response is assigned the value `0`.
- Formula for each selection: `([Highest normalization input value for the selected responses] / [Maximum of the normalization input values]) * 100`.
- The score for the metric \(question\) is the maximum calculated score among all responses.
In this example, the user selects **Dog** and **Cat**.
- The score for the **Dog** response is `(0 / 9) * 100 = 0`.
- The score for the **Cat** response is `(5 / 9) * 100 = 55.5`.
- The score for the overall metric is `55.5`.
Formula: `(Highest of the normalization input values for the selected responses / Highest of all the normalization input values) * 100`.
In this example, the question allows for multiple selections among answers with values of `1`, `2`, and `4` and normalization input values of `3`, `5`, and `9` respectively.
|Response|Calculation|Score|
|--------|-----------|-----|
|Dog and Cat|\(5 / 9\) \* 100|56|
|Cat|\(5 / 9\) \* 100|56|
|Cat and Goldfish|\(9 / 9\) \* 100|100|
|Goldfish|\(9 / 9\) \* 100|100|