Home  Labs 
Lab
Stroop Effect
Preparation
Reading: Chapter 2/How Psychologists Do Research, Chapter 6/Sensation and Perception
HandsOnPsych: Research Methods and Statistics, Perception [The data for this study will be collected from within this module]
Live!Psych: Module 2.3, How Psychologists Do Research/The Experimental Method
Introduction
The unique, and valuable, characteristic of an experiment is that you can potentially learn something about what causes what. You start with a theory or explanatory model that states an explanation of how variable A can affect variable B if the theory is correct. Variable A is the independent variable and variable B is the dependent variable. Variable B is what you measure and Variable A is the condition that you manipulate. So B is "dependent" on A. In this experiment we will use a simple test of how quickly the subjects (Ss) can read a list of words as the dependent variable.
In the HandsOnPsych material on the CDROM in the Module on Perception, Part One is called the Principles of Perception. In this section is a demonstration of what is called the "Stroop Effect." Do this demonstration by yourself. This is a simple demonstration of how the brain works more slowly when it is asked to process two pieces of information that are contradictory. It provides you with an automated procedure that allows for the collection of data and lends itself to the design of an experiment that uses reading time as a dependent variable.
Design an experiment in which each of your subjects does this procedure under two of the conditions: word colour match, and word colour mismatch, which will be your independent variable. You will need a theory that explains why your independent variable should affect the speed of your subjects' reading. If you would like to read a bit about Stroop's thoughts on the theory have a look at the one of the following:
Neuroscience for Kids—The Stroop Effect
Interference: The Stroop Effect
NOVA Online Everest: Introduction to the Stroop Test
Research on the Stroop Effect [Pretty advanced]
Or read Stroop's original publication:
Procedure:
In the Perception module of HandsOnPsych there is a section on the "Principles of Perception." Work your way through this module until you come to the demonstration called the "Stroop Effect." This little demonstration is set up in such a way that it contains three levels of an independent variable and a dependent variable. The three levels or conditions of the independent variable are 1) coloured nonsense words, 2) words the same colour as the word (wordcolour match), and 3) words a different colour than the word (wordcolour mismatch) . In your experiment you will use just the last two conditions.
Your theory is your explanation as to why and how your variable should affect memory. Your hypothesis is your specific prediction about the outcome of the experiment. "Condition # 1 (Wordcolour match) will have an average reading time greater or less than condition # 2 (wordcolour mismatch)."
Alternate Procedure: If you don't have access to computers with HandsOnPsych on it there are a couple of other ways to do the Stroop experiment. See the links above.
Data Collection:
Record your data in a chart like the one below:
Subject number 
Control Group or Condition 1 wordcolour match 
Test Group or Condition 2 wordcolour mismatch 
Subject 1 


2 


3 


4 


5 


etc. 


Results:
Your results will be known when you calculate a ttest that compares the two groups. You can either use VassarStats web site or an Excel spreadsheet.
Calculations:
The way to test for the difference between two groups or groups of numbers where there is an independent variable and a dependent variable is with a "t" test. To calculate a t statistic and the p value associated with it you can use the VassarStat site:
To use VassarStat go to this statistics web site. http://faculty.vassar.edu/lowry/VassarStats.html
Click on "ttests and procedures." Then "ttest of independent samples." You should arrive here: [ http://faculty.vassar.edu/lowry/VassarStats.html ]
You will first be asked to enter the size of your sample. Remember the larger the sample, the better test it is, but 10 or so should more than enough. Then you will find a data entry chart the size of your sample where you can enter the values for condition 1 and condition 2 or X and Y as they see it. Then click on calculate and you will find out the results of your experiment. Below the table you will then find the t value and p value. Remember that the "p" value is the probability that any difference between your two groups is due to chance and not the effect of your independent variable.
Test of the VassarStat system. If you enter the following data
Subject number 
Control Group or Condition 1 X (wordcolour match) 
Test Group or Condition 26 Y (wordcolour mismatch) 
Subject 1 
5 
9 
2 
7 
12 
3 
6 
10 
4 
4 
11 
5 
5 
8 
6. 
5 
15 
To test that you are using VassarStats correctly, you can enter the following data and compare your results with what has been previously calculated.
When you are confident that you know how to use the system, enter your own data.
Your calculations from VassarStats should agree with these:
VassarStats
Printable Report
tTest for Correlated Samples
Date
Values
entered:
count 
X_{a}

X_{b}

X_{a}
 X_{b} 
1 
5 
9 
4 
Summary
Values
Values

X_{a}

X_{b}

X_{a}
 X_{b} 
n 
6 
6 
6 
sum 
32 
65 
33 
mean 
5.3333

10.8333

5.5 
sum_sq

176 
735 
215 
SS 
5.3333

30.8333

33.5 
variance

1.0667

6.1667

6.7 
st.
dev. 
1.0328

2.4833

2.5884 
Variances
and standard deviations are calculated
with denominator = n1.
Mean_{A}
 Mean_{B}

t 
df 
5.5 
5.2 
5 
P 
onetailed

0.0017265

twotailed

0.003453 
Excel Calculations:
Open a new spreadsheet.
Enter your data, using the same data that you did above.
Click on a cell below the column with the Variable 1 or X data and type "p="
Click in the cell next to the p= and click on the f_{x }icon_{ }or click on insert "function." Choose "Statistical" for Function Category and "TTEST" for Function name.
You will be asked to fill four boxes. Click in the box to the right of "Array1" and then select the column with the data for group X or 1. Then click in the box to the right of Array2 and then select the numbers or data associated with group Y or 2. This is a "one tailed" test because you predicted which group would have the higher score than the other so put a "1" to the right of "tails." This is a "paired" set of data because you used one group of subjects using them in both conditions, this is also called a "within subjects" design, so put a "1" to the right of Type. then click on "OK" and a p value for the t statistic should appear in the cell you selected above. This number should be the same as the "p" value for the onetailed test that you calculated at the VassarStats site.
When you have collected your results in a table similar to the one below enter your data into VassarStats, an Excel spreadsheet or both and do the calculations in the same manner as you did in the test run.
Reporting:
Create a file with the word processor program that you normally use. Label this file with your name and "Psychology Lab One." Save this file as a "Rich Text Format" (.rtf) file.
Provide the following information in your file:
Name:
Course:
Instructor:
Date:
Location of Observations:
Number of subjects observed and their general nature (female vs. male for example).
The theory you are testing:
Your hypothesis: (Basically which condition will produce the better memory results)
Summary of observations: (What did you observe?)
Fill in the data chart:
Subject number 
Control Group or Condition 1 (wordcolour match) 
Test Group or Condition 2 (wordcolour mismatch) 
Subject 1 


2 


3 


4 


5 


etc. 


Mean 


Standard Deviation 


t =
one tailed p =
If your "p" value is less than .05 we will accept the hypothesis.
Conclusions: What do your results tell you about the accuracy or validity of your theory?
Email this file to your instructor using the email or bulletin board system requested for this assignment.