How to Calculate Effect Sizes from Published Research: A Simplified Spreadsheet

By Will Thalheimer and Samantha Cook

Note: This spreadsheet follows the same outline as the article by Thalheimer and Cook (2002).
How to Calculate Effect Sizes from Published Research: A Simplified Methodology

Updated April 2003 to deal more effectively with situations where the treatment-group mean is lower than the comparison-group mean.

TABLE OF CONTENTS

Sheet 1 -- Calculating Cohen's d from t-test

Sheet 2 -- Calculating Cohen's d from t-test when you don't have Standard Deviations or Standard Errors
Sheet 3 -- Calculating Cohen's d from t-test when you have Standard Errors instead of Standard Deviations

Sheet 4 -- Calculating Cohen's d from F-test

Sheet 5 -- Calculating Cohen's d from F-test when you don't have MSE's

Calculating Cohen's d from t-tests

Subjects Mean Standard Deviation

Treatment Condition 16 1.121 0.07 Input Appropriate Numbers

Comparison Condition 17 1.26 0.08 in the cells to the left!!

Cohen's d 1.9 huge effect Relative Size of Cohen's d

negligible effect (>= -0.15 and <.15)
numerator -0.139 small effect (>=.15 and <.40)
den's num 1st 0.0735 medium effect (>=.40 and <.75)
den's num 2nd 0.1024 large effect (>=.75 and <1.10)
den's denominator 33 very large effect (>=1.10 and <1.45)
den's total (s pooled) 0.0730089243 huge effect >1.45

Percent Change Relative Size of % Change

(From Comparison to Treatment) -11
small decrease huge decrease <-75
very large decrease (<=-50 and >-75)
large decrease (<=-30 and >-50)
medium decrease (<=-15 and >-30)
small decrease (<=-5 and >-15)
negligible change (>= -5 and <5)
small increase (>=5 and <15)
medium increase (>=15 and <30)
large increase (>=30 and <50)
very large increase (>=50 and <75)
huge increase >75
Spreadsheet developed by
by Will Thalheimer (Work-Learning Reseach)
and Samantha Cook (Harvard University)

Cohen's Effect Percentile Percent of

Standard Size Standing Nonoverlap
2 97.7 81.10%
1.9 97.1 79.40%
1.8 96.4 77.40%
1.7 95.5 75.40%
1.6 94.5 73.10%
1.5 93.3 70.70%
1.4 91.9 68.10%
1.3 90 65.30%
1.2 88 62.20%
1.1 86 58.90%
1 84 55.40%
0.9 82 51.60%
LARGE 0.8 79 47.40%
0.7 76 43.00%
0.6 73 38.20%
MEDIUM 0.5 69 33.00%
0.4 66 27.40%
0.3 62 21.30%
SMALL 0.2 58 14.70%
0.1 54 7.70%
0 50 0
TABLE OF CONTENTS

Sheet 1 -- Calculating Cohen's d from t-test

Sheet 2 -- Calculating Cohen's d from t-test when you don't have Standard Deviations or Standard Errors
Sheet 3 -- Calculating Cohen's d from t-test when you have Standard Errors instead of Standard Deviations

Sheet 4 -- Calculating Cohen's d from F-test

Sheet 5 -- Calculating Cohen's d from F-test when you don't have MSE's

Sheet 2 -- Calculating Cohen's d from t-tests

when you don't have Standard Deviations or Standard Errors

Treatment Condition
d=t
Subjects
230
(
nt + n c n t + n c
nt nc nt + nc 2 )( t statistic
4.731
)
Input Appropriate Numbers
in the cells to the left!!
Comparison Condition 230

Cohen's d 0.44 medium effect Relative Size of Cohen's d

negligible effect (>= -0.15 and <.15)
1st numerator 460 small effect (>=.15 and <.40)
1st denominator 52900 medium effect (>=.40 and <.75)
2nd numerator 460 large effect (>=.75 and <1.10)
2nd denominator 458 very large effect (>=1.10 and <1.45)
Within Square Root 0.0934538627 huge effect >1.45
TABLE OF CONTENTS

Sheet 1 -- Calculating Cohen's d from t-test

Sheet 2 -- Calculating Cohen's d from t-test when you don't have Standard Deviations or Standard Errors
Sheet 3 -- Calculating Cohen's d from t-test when you have Standard Errors instead of Standard Deviations

Sheet 4 -- Calculating Cohen's d from F-test

Sheet 5 -- Calculating Cohen's d from F-test when you don't have MSE's

Sheet 3 -- Calculating Cohen's d from t-tests

when you have Standard Errors instead of Standard Deviations

x t x c
d=
( nt 1) SE t n 2 + ( nc 1) ( SE c n c )2

Subjects
( t

n t + nc
)

Mean Standard Errors

Treatment Condition 24 57 0.226 Input Appropriate Numbers
Comparison Condition 24 53 0.272 in the cells to the left!!

Cohen's d 3.34 huge effect Relative Size of Cohen's d

negligible effect (>= -0.15 and <.15)
numerator 4 small effect (>=.15 and <.40)
den's num 1st 28.193952 medium effect (>=.40 and <.75)
den's num 2nd 40.839168 large effect (>=.75 and <1.10)
den's denominator 48 very large effect (>=1.10 and <1.45)
den's total (s pooled) 1.1992455962 huge effect >1.45

Percent Change negligible change Relative Size of % Change

(From Comparison to Treatment) 8
huge decrease <-75
very large decrease (<=-50 and >-75)
large decrease (<=-30 and >-50)
medium decrease (<=-15 and >-30)
small decrease (<=-5 and >-15)
negligible change (>= -5 and <5)
small increase (>=5 and <15)
medium increase (>=15 and <30)
large increase (>=30 and <50)
very large increase (>=50 and <75)
huge increase >75
TABLE OF CONTENTS

Sheet 1 -- Calculating Cohen's d from t-test

Sheet 2 -- Calculating Cohen's d from t-test when you don't have Standard Deviations or Standard Errors
Sheet 3 -- Calculating Cohen's d from t-test when you have Standard Errors instead of Standard Deviations

Sheet 4 -- Calculating Cohen's d from F-test

Sheet 5 -- Calculating Cohen's d from F-test when you don't have MSE's

Sheet 4 -- Calculating Cohen's d from F-tests

x t x c
d=

Subjects
MSE

Mean
(
nt +n c2
n t +n c )
MSE
Treatment Condition 48 1.69 2.34 Input Appropriate Numbers
Comparison Condition 48 4.64 in the cells to the left!!

Cohen's d 1.95 huge effect Relative Size of Cohen's d

negligible effect (>= -0.15 and <.15)
numerator -2.95 small effect (>=.15 and <.40)
den's numerator 94 medium effect (>=.40 and <.75)
den's denominator 96 large effect (>=.75 and <1.10)
denominator total 1.5136875503 very large effect (>=1.10 and <1.45)
huge effect >1.45

Percent Change negligible change Relative Size of % Change

(From Comparison to Treatment) -64
huge decrease <-75
very large decrease (<=-50 and >-75)
large decrease (<=-30 and >-50)
medium decrease (<=-15 and >-30)
small decrease (<=-5 and >-15)
negligible change (>= -5 and <5)
small increase (>=5 and <15)
medium increase (>=15 and <30)
large increase (>=30 and <50)
very large increase (>=50 and <75)
huge increase >75
TABLE OF CONTENTS

Sheet 1 -- Calculating Cohen's d from t-test

Sheet 2 -- Calculating Cohen's d from t-test when you don't have Standard Deviations or Standard Errors
Sheet 3 -- Calculating Cohen's d from t-test when you have Standard Errors instead of Standard Deviations

Sheet 4 -- Calculating Cohen's d from F-test

Sheet 5 -- Calculating Cohen's d from F-test when you don't have MSE's

Sheet 5 -- Calculating Cohen's d from F-tests

when you don't have MSE

Treatment Condition
d= F
Subjects
22
(
nt + nc nt + nc
nt nc nt + nc 2 )(
F statistic
8.39
)
Input Appropriate Numbers
Comparison Condition 33 in the cells to the left!!

Cohen's d 0.81 large effect Relative Size of Cohen's d

negligible effect (>= -0.15 and <.15)
1st numerator 55 small effect (>=.15 and <.40)
1st denominator 726 medium effect (>=.40 and <.75)
2nd numerator 55 large effect (>=.75 and <1.10)
2nd denominator 53 very large effect (>=1.10 and <1.45)
Within Square Root 0.8121521994 huge effect >1.45