Tables for Determining Approximate Sample Size

Table 1.  Approximate Sample Size Required
for Testing a Hypothesis about a Single Mean
                             
                             
          Beta =.05         Beta =.10        Beta =.20
                
  d    a = .05  a = .01  a= .05  a = .01 a = .05  a = .01
                
                          Two-Tailed Test
                                        
1.50     8         12       7       10       6        9
1.00    15         22      13       19      10       15
 .75    25         36      21       31      16       25
 .50    55         74      43       64      34       51
 .25   208        285     168      238     126      187
                                        
                         One-Tailed Test
  
1.50     7         10       6        9      5         7
1.00    13         19      10       16      8        13
 .75    22         31      17       26     11        21
 .50    46         65      36       54     27        42
 .25   173        252     137      208     99       161
  
  
Table 2. Approximate Sample Size Required for Testing
a Hypothesis about between Two Independent Means-a
  
          Beta =.05         Beta =.10        Beta =.20
                
  d    a = .05  a = .01  a= .05  a= .01  a = .05  a = .01
  
  
                         Two-Tailed Test

1.50      14       20       12       17       9       14
1.00      28       40       23       34      18       27
.75       48       69       40       58      31       46
.50      106      150       87      125      66      100
.25      416      570      336      476     251      374
                                        
                          One-Tailed Test
  
1.50      11       17       9        15       7       12
1.00      23       34      19        29      14       23
 .75      40       59      32        50      23       38
 .50      90      130      71       108      51       85
 .25     346      505     274       417     198      321
  
a The tabled values indicate requires size for each of the two samples.  It is assumed that standard deviation of x scores is equal to the standard deviation of y scores.