What Is Recode?
Recode is a feature used when you want to load a single variable and change its label (questionnaire) or choice text, or group multiple choices together according to your desired criteria. For example, if a question asking about place of residence has the choices 'Seoul,' 'Gyeonggi-do,' and 'Incheon' separately, you can group them into a single choice called 'Capital area' to conduct analysis more smoothly according to your desired criteria.
How to Recode
Recode methods are described separately according to the type of recode being used.
📌 The create variable feature is available on the Professional or Enterprise plan.
Creating a New Variable by Grouping Choices
1️⃣ Select a variable on the Recode screen.
step 1. Click [Variables] at the top of the Analytics screen, then click [Create variable > Recode] from the left menu.
step 2. Click the target variable dropdown bar and select the 'target variable' for the recode operation.
2️⃣ Group choices.
step 1. In the target variable work area, select the values to group together and click the [Group] button.
step 2. The grouped values appear in the 'New variable' on the right. You can edit the variable name, variable label, and new value label of the grouped values as desired, and select an appropriate variable type based on the target variable.
step 3. Group or [Copy] the remaining values according to your desired criteria, then edit the new value labels.
💡 Usage tip | If you want to change the order of values in the new variable, use the [Remove] button to remove the value, then regroup it in the desired order.
4️⃣ Create the new variable.
step 1. Once you've finished configuring the new variable settings, click [Create variable] in the upper right.
step 2. Check the settings in 'New variable preview,' then click [Done].
step 3. The newly created variable can be viewed in the 'Variable list' and in the 'Results tab,' 'Cross-tab tab,' and 'Report tab' of Analytics.
Creating a New Variable by Changing the Name, Label, Type, and Value Labels of an Existing Variable
1️⃣ Changing the variable name
In variable recode, after selecting the target variable, you can edit the 'Variable name' in the [New variable] section on the right to your desired name.
Variable names can use English letters, numbers, and some special characters (_, -, ., (), ~, {}, [], +).
💡 Usage tip | A name that duplicates an existing variable name cannot be entered.
2️⃣ Changing the variable label
In variable recode, after selecting the target variable, you can edit the 'Variable label' in the [New variable] section on the right to your desired label. The label of the target variable is automatically entered as the default, and Korean, English, numbers, and special characters can all be used.
3️⃣ Changing the variable type
In variable recode, after selecting the target variable, you can select your desired type from the 'Variable type' in the [New variable] section on the right. The selectable types are displayed differently depending on the type of the target variable.
🙋 When and how should you change the variable type?
You can change the variable type depending on the analysis purpose or how you want to view the data. Refer to the examples below to select the appropriate type for your situation.
✅ When the current type is single choice
This is for cases of transforming a multiple-choice question designed for selecting only one option from the start. (e.g., place of residence, preferred brand)
New variable type | Use this when | Variable type change example |
Single choice | When there are too many choices making graphs or results hard to view, or when you want to group multiple choices into larger categories | Group regions like Seoul / Gyeonggi / Incheon / Gangwon / Chungbuk / Chungnam → Capital area / Chungcheong area / Gangwon area to create new choices. |
Multiple choice | When you want to simultaneously classify a single response under multiple criteria | For example, transform a response selecting 'iPhone' so that it is included in both the Apple (manufacturer) group and the Smartphone (product type) group. |
Ranking | When you want to sort response items in a specific order to suit your analysis purpose | Sort a randomly listed set of brands according to your desired criteria such as market share order or preference order. |
✅ When transforming rating (scale) data
This is for cases of transforming data collected with scores like "Very satisfied (5 points) ~ Very dissatisfied (1 point)." (e.g., service satisfaction, design preference)
New variable type | Use this when | Variable type change example |
Single choice | When you want to group rating scales into a few categories for more intuitive viewing | Group scores from a 5-point scale: 4–5 points as 'Satisfied,' 3 points as 'Neutral,' 1–2 points as 'Dissatisfied.' |
Numeric | When you want to convert rating responses to numbers for calculating averages or statistics | Convert text responses to numbers: 'Very satisfied' → 5, 'Satisfied' → 4, 'Neutral' → 3. |
Rating | When you want to convert the scoring system to a different scale | Convert a 5-point scale result to a 100-point basis for comparison with other survey results on the same standard. |
✅ When transforming numeric data
This is for cases of transforming data entered directly as numbers like age (23, 45) or amounts (15,000). (e.g., age, income, usage time)
New variable type | Use this when | Variable type change example |
Single choice | When you want to divide numeric data into intervals for group-based analysis | Create age intervals: 21, 25, 29 → 20s; 31, 35 → 30s. |
Numeric | When you want to change the unit of entered numbers or correct incorrectly entered values | Convert a large amount like 1,000,000 won to 100 (in units of 10,000 won), or correct incorrectly entered values. |
4️⃣ Changing value labels
step 1. In variable recode, after selecting the target variable, you can directly edit to your desired value label in the 'New value label' field in the [New variable] section on the right.
step 2. Clicking the [Edit all at once] button lets you edit all value labels at once. When done, click [Save].
Recode Operations by Variable Type
1️⃣ Numeric → Single choice recode (auto grouping)
When recoding a numeric variable to a single-choice variable, you can use the auto grouping feature.
💡 Usage tip | Auto grouping is a feature that automatically divides numeric responses into intervals and converts them into choice values.
step 1. Select a numeric variable from 'Target variable.'
step 2. In the 'New variable' area, enter the variable name and variable label, then set the variable type to 'Single choice.'
step 3. Click the [Auto group] button, set the minimum, maximum, and interval values, then click [OK].
Minimum: Sets the minimum value of the range to be converted to choice values from among numeric responses. For example, if the purchase price of wireless earbuds was collected in 'won' as a numeric value, values smaller than 10,000 won may be erroneous responses, so you can set the minimum to 10,000.
Maximum: Sets the maximum value of the range to be converted to choice values from among numeric responses. For example, if the usage period of wireless earbuds was collected in 'months,' values like 50 or 100 months could be erroneous responses, so you can limit the maximum appropriately.
Interval: Sets the interval at which responses from the minimum to maximum will be divided. For example, setting a minimum of 1 month, maximum of 36 months (3 years), and interval of 6 months for wireless earbud usage period would automatically create 5 choice values: 1–6 months, 7–12 months, 13–18 months, 19–24 months, 25–30 months, and 31–36 months.
step 4. Single choice values are automatically generated in the new variable work area. Edit the automatically entered value labels to your desired criteria, then click [Create variable].
2️⃣ Numeric to Numeric recode
When recoding a numeric variable back to numeric, you can use the value replacement feature. Value replacement is used when you want to replace a specific response value with a different value.
🧙 Understanding value replacement through an example
For example, suppose wireless earbud usage period was collected as a numeric value in 'months,' but a respondent entered a value like 360 (30 years) due to misunderstanding the unit.
You can clean the data by replacing values not suitable for analysis with the overall average or another value.
step 1. Select a numeric variable from 'Target variable,' then set the variable type to 'Numeric' in the 'New variable' area.
step 2. Select the value to replace on the left, then click the [Replace value] button. Then select your preferred method from the following options.
Average: Replace with the overall numeric response average for that variable.
Non-response: Treat the response as non-response.
0: Replace the response with the number 0.
Custom input: Replace with a value you enter directly, other than average, non-response, or 0.
step 3. After completing the value replacement, copy the remaining values and click [Create variable].
💡 Usage tip | In numeric recode, 'value replacement' and 'auto grouping' cannot be used simultaneously. To use both features, first apply 'value replacement' to create a numeric variable, then recode the created variable to a choice type.
Have you been making good use of the recode feature?
If you still have questions after reviewing this guide, please contact us anytime via the [Customer Support icon] in the bottom right corner of your screen. Our team will do its best to help you resolve any difficulties you're experiencing.