Setting Variables with Levels of Measurement: Discover Statistics with JASP for Beginners (3 of 6)

Introduction to JASP Data Manipulation

JASP allows for effective data manipulation through the setting of levels of measurement and value labels once we have obtained a dataset. This process is essential in understanding our data for statistical analysis.

Data Formats in JASP

  • JASP Files: Contain preset variable information.

  • Raw Data Files: Often in CSV format and require setting up variables.

    • To open a dataset:

      1. Click on the Main Menu icon.

      2. Select Open > Data Library.

      3. Choose the debug data set, identifiable by the lack of a JASP logo.

Understanding Variable Types in JASP

JASP recognizes different types of categorical variables, each represented by specific icons:

  • Nominal Variables:

    • Icon: Three balls.

    • Variables coded as categories (e.g., 1 for experimental, 0 for control).

  • Nominal Text Variables:

    • Icon: Three balls with a small "a."

    • Similar to nominal, but data is entered as text (e.g., Gender as 'M' and 'F').

  • Ordinal Variables:

    • Icon: Three bars.

    • Reflect an order (e.g., severity of diabetes).

  • Continuous Numeric Variables (Scale Data):

    • Icon: Yellow ruler.

    • Represents numeric data, often with decimal points and no inherent labels.

Setting Value Labels

  • Categorical variables allow setting value labels that will reflect in analysis results.

    1. Click on the column name to open a settings box.

    2. Edit labels by double-clicking on the value and typing the desired label (e.g., "Control" for 0).

    3. Changes automatically update in corresponding tables.

Reordering Value Labels

  • Value labels can be reordered for clarity:

    • Select label and use arrows to adjust positions or reverse the entire order for efficiency.

Changing Variable Types

JASP allows for changing variable types based on data needs:

  • You can change scale variables to nominal when they represent categories rather than measurements.

  • Be cautious: Changing variable types may result in loss of labels.

    • Example: V1 (participant ID) initially read as scale can be redefined as nominal.

Constraints in Variable Types

  • Some variables cannot be altered due to the nature of their data:

    • If a variable is identified as a scale, it might not convert to ordinal without data errors.

    • Nominal text variables remain as text and cannot change type.

Importance of Setting Variable Levels

Setting levels in JASP is beneficial for several reasons:

  1. Facilitates Variable Selection: Helps in organizing variables correctly for statistical tests.

  2. Encourages Data Familiarity: Promotes thorough data examination, highlighting potential issues pre-analysis.

  3. Enhances Data Visualization: Enables JASP to generate appropriate graphs based on variable types.

Conclusion

  • For practice, the debug dataset can be closed without saving. The focus of this guide was on setting variable levels rather than the dataset's content. Future instructions will utilize a more meaningful dataset to demonstrate real-world variable handling.