PSYC 2520 Chapter Notes - Chapter 13: Bias Of An Estimator, Null Hypothesis, Statistical Significance

• School: York University
• Department: Psychology
• Course: PSYC 2520
• Professor: Josee Rivest

Document Summary

If you are interested in comparing the means of two groups, you will find the t test a convenient and powerful tool. It will allow you to test the likelihood that the population means represented by the two groups are equal, by setting up a signal-to-noise ratio. In this ratio, the signal is represented by the difference between the two means, and the noise is represented by the variability of the scores within the samples. The larger the signal is relative to the noise, the more likely it is that the null hypothesis will be rejected. When we compare the mean differences between groups, we should also take into consideration the amount of variability within the groups. Larger t values have a lower level of probability. When we want to compare two independent samples, a general-purpose formula for the independent-sample t is t = m1 m2 / (1 / n1 + 1 / n2)s2.