What Is Reverse Scoring?
Reverse scoring is the process of flipping the value of a response on a Likert-type scale so that the direction of the item aligns with other items measuring the same construct.
For example, on a 5-point scale:
Original Response
|
Reverse-Scored Value
|
1 (Strongly Disagree)
|
5
|
2
|
4
|
3
|
3
|
4
|
2
|
5 (Strongly Agree)
|
1
|
Why Is Reverse Scoring Used in Validated Instruments?
- Controlling for Response Bias
Including both positively and negatively worded items helps reduce acquiescence bias (the tendency to agree with all statements) and inattentive responding.
- Enhancing Construct Validity
Reverse-worded items help ensure that participants are reading carefully and not simply responding based on item format.
- Balanced Item Design
A well-constructed scale includes items that frame the construct from both positive and negative perspectives. For example, in a self-esteem scale:
- Positive item: “I feel that I have a number of good qualities.”
- Negative item: “I feel I do not have much to be proud of.”
The negative item would need to be reverse-scored to match the positive direction of the scale.
How to Reverse Score in SPSS
Let’s say you have a 5-point Likert scale (1 = Strongly Disagree, 5 = Strongly Agree), and Item Q5 is negatively worded.
Step-by-Step in SPSS:
- Identify Reverse-Coded Items
This is usually specified in the instrument manual or codebook.
- Create a New Variable for the Reverse-Scored Item
- Go to Transform > Compute Variable
- In the “Target Variable” box, type a new name (e.g., Q5_rev)
- In the “Numeric Expression” box, enter the formula:
6 - Q5
(This formula flips the values for a 1–5 scale.)
- Click OK
SPSS will create a new variable (Q5_rev) with the reverse-coded values.
- Use the Reverse-Scored Variable in Analysis
When computing scale totals or averages (e.g., in Transform > Compute Variable or Analyze > Scale > Reliability Analysis), use the reverse-scored variable, not the original.
Example: Computing a Total Score
Suppose you have a 5-item scale:
- Q1, Q2, Q3 (positively worded)
- Q4, Q5 (negatively worded)
You reverse-code Q4 and Q5 as Q4_rev and Q5_rev.
To compute the total score:
Use Q1 + Q2 + Q3 + Q4_rev + Q5_rev
Important Notes for Doctoral Students:
- Do Not Reverse the Text—Only the Scoring
The item remains phrased negatively; only the numerical response is reversed.
- Document Everything
In your methodology chapter, clearly indicate which items were reverse scored, why, and how.
- Check Reliability Before and After
Cronbach’s alpha can vary if reverse-coded items are not handled correctly. Always check the internal consistency of your scale with reverse-scored items properly transformed.