Implementation of an Instrument

College of Doctoral Studies

RES-842 Reverse Coding Procedure Using SPSS

By Dr. Alex Casteel

Introduction

When developing instruments, researchers will test a construct by asking the participant to judge the participant’s response about a characteristic. For example, a test of extraversion would ordinarily seek out one’s preference for multiple stimuli, social gatherings, and comfort with relationships. High scores in response to these types of items indicate higher manifestations of trait extraversion. An additional approach to this technique is to ask questions about the low end of the dimension, in this case, low extraversion, or introversion. The item may ask one’s preference for quiet environments in which a high score indicates that preference. However, to use that score to indicate one’s preference for trait extraversion, the score must be manipulated. The manipulation is completed by reverse scoring. Reverse scoring is completed by changing the values of strong preference for a reciprocal construct (e.g., introversion) to a low score for the primary construct being tested (e.g., extraversion). The result is that the score for the instrument reflects the overall measure of the individual for the primary construct (e.g., extraversion).

Requirements

The instrument being used must identify which items are to be reverse coded. Instrument instructions will often identify these items by placing an R after the item number. For example, in the following sequence—1, 2R, 3, 4R, 5R—items 2, 4, and 5 are reverse coded. Reverse coding is a simple arithmetic process of subtracting the score for the item from a value of ( k + 1) in which k is the number of response choices on a Likert scale. For example, if the scale measure one’s preference from 1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, and 5 = strongly agree, the number of items is five, so k = 5. To reverse code a score, X, on a 5-item Likert scale, one uses the formula ( k + 1) – X. The results of this operation are shown in Table 1.

Table 1.

Example of Reverse Coding for an Item Using a 5-Item Likert Scale

Item score for reverse coded item ( X)	Application of formula ( k + 1) – X	Result
1	(5 + 1) – 1	5
2	(5 + 1) – 2	4
3	(5 + 1) – 3	3
4	(5 + 1) – 4	2
5	(5 + 1) – 5	1
Note. k = 5 for 5-item Likert scale.

SPSS offers the ability to reverse code instrument items. The following example will show how to reverse code three items that use the same Likert scale.

Notes

SPSS can recode all the reverse coding items in one step provided all use the same scale (e.g., 1 – 5). This process will replace old values with the reverse coded values. SPSS refers to these as Variables; however, technically speaking, you are recoding instrument items. The variables are calculated from the associated items. For this example, the SPSS term of Variable is used to remain consistent with the SPSS prompts.

<Document continues on next page.>

Steps

1. Select Transform > Recode into Same Variables…

Figure 1.

SPSS Transform Menu to Select Recode Into Same Variable

<Document continues on next page.>

2. Select the items to be recoded and move them into the variables area. The figure below shows items E2, E4, and E5 moved into the variables area.

Figure 2.

Recode Into Same Variables Wizard

<Document continues on next page.>

3. Select Old and New Values…

Figure 3.

Recode Into Same Variables: Old and New Values Wizard

<Document continues on next page.>

4. Input the scores as they should be recoded. For example, an old value 1 now becomes a 5. Input those values, and then select Add, which will move that relationship to the window below.

Figure 4.

Inputting Old and New Values Example

<Document continues on next page.>

Figure 5.

Completing Input of Old and New Values for One Variable

<Document continues on next page.>

5. Repeat this process for all values to be recoded. Note that in the case of an odd numbered Likert scale, the middle value does not change—a value of 3 remains 3, so it will not be recoded.

Figure 6.

Complete Algorithm for Recoding Variables

<Document continues on next page.>

6. Once you confirm all values to be recoded are correctly represented in the window, select Continue, which will take you back to the Recode into Same Variables window, but now the OK button is available. To complete the recoding, select OK.

Figure 7.

Completing Reverse Coding of Variables

7. You have successfully reverse coded the Variables (instrument items) from your instrument in SPSS. The Variables (items) keep the same Variable Name in SPSS (e.g., E2, E4, and E5 in the example). You may now use that item to calculate the participant scores.

© 2022. Grand Canyon University. All Rights Reserved.

© 2022. Grand Canyon University. All Rights Reserved.