Below I've outlined the steps to create a new variable that expresses whether the response of which color was shown matched the color actually shown in the experiment:

Note that you cannot include the IF function directly in the Numeric Expression box, but you can use the IF button to open the If dialog box and enter the condition there. Alternatively, you can use the following formula in the Numeric Expression box to achieve the same result:

response_color = actual_color

This formula returns a logical value (TRUE or FALSE) depending on whether the values in the two variables match. You can then convert this logical value to a numeric value (1 or 0) by wrapping the formula in the following function:

NUM(response_color = actual_color)

Hope this helps.

Hey, thank you for the detailed answer ! I tried to apply this to create a variable that detects and marks outliers within the variable response time. In this case outliers are below 200ms and above 1200 ms. I entered the correct if function (rt1200), and the numerical expression 1. But it keeps giving me dots in the created variable in the data view. Chat GPT couldn't resolve this either. I'm still very new to SPSS, as you may have noticed . But if you instantly notice the error, please let me know. Would be much appreciated. Thanks very much in any case !

Kilian Meller At first glance, it I suspect that this is a result of the numerical expression of 1. This should actually be the value that the new variable takes on when the condition is true (i.e., when the response time is an outlier). In this case, the value should be "1".

Instead, you can create a variable that detects and marks outliers within the variable response time:

This formula checks whether the value in the rt variable is less than 200 or greater than 1200. If the value meets this condition, it returns a 1, indicating that the value is an outlier. If the value does not meet this condition, it returns a 0. Thus, it may not be fixed at "1".

Note that the dots in the created variable in the data view indicate missing values. If there are missing values in the rt variable, they will also appear as missing values in the new variable.

Let me know how this works.

Thank you, I asked my prof. this, and she said I cannot or should not put IF into the numerical expression box. But it may be possible in your way too. But now I got the variable such that it puts 1 for outliers. All I need to do now is have it put 0 instead of dots for values in the normal range. I know I need to use ELSE for it, but the exact expression I didnt find. I will leave this for now, as it takes too much time fiddling with these expression, but the gist of it I got I think. I did not find a tutorial specifically on these very expressions on yt. Please let me know if you know one. Thank you very much for your help in any case ! Much appreciated!

Kilian Meller If I am understanding correctly, it sounds like you are trying to replace the dots (which represent missing values) with 0s for the data that falls within the normal range. This can be done with the COMPUTE command in SPSS and include an IF-ELSE statement.

The general syntax would be:

COMPUTE new_variable = IF old_variable < lower_limit OR old_variable > upper_limit OR MISSING(old_variable) THEN 1 ELSE 0.

• "new_variable" is the name you want to give to the new variable
• "old_variable" is the name of the original variable that you are evaluating
• "lower_limit" and "upper_limit" are the minimum and maximum values that define the normal range
• "1" and "0" represent the values that will be assigned to the new variable based on the IF-ELSE statement.

For example, if you want to replace the dots with 0s for values between 10 and 80 in a variable called "score", you can use the following syntax:

COMPUTE new_score = IF score < 10 OR score > 80 OR MISSING(score) THEN 1 ELSE 0.

Hope this helps.