Standard Article

Hypothesis Testing Related to Differences Parametric Tests

Part 2. Marketing Research

  1. Marcus J. Schmidt

Published Online: 15 DEC 2010

DOI: 10.1002/9781444316568.wiem02023

Wiley International Encyclopedia of Marketing

Wiley International Encyclopedia of Marketing

How to Cite

Schmidt, M. J. 2010. Hypothesis Testing Related to Differences Parametric Tests. Wiley International Encyclopedia of Marketing. 2.

Author Information

  1. Copenhagen Business School, Copenhagen, Denmark

Publication History

  1. Published Online: 15 DEC 2010

Abstract

This article covers parametric tests of one and two samples. First, the implications of making decisions under uncertainty are discussed. Next, the concepts of hypothesis testing, probability values, and confidence intervals are introduced, and the steps of the hypothesis testing procedure are explicated. The consequences of making false decisions (type I and type II errors) are clarified and the difference between a one- and a two-tailed test is made clear. The following tests are covered: large sample test (i) of the mean and (ii) of the population proportion, and small sample test of (iii) mean and (iv) proportion. First, we show how to conduct the tests based on one sample where the empirical dataset is compared to a hypothesized value. Thereafter, we repeat the same four test conditions but now the focus is on the difference between two empirical samples. In separate sections, we describe how to carry out a matched sample test, a test for equality of two population variances, and a test of difference between means when variances are not equal.

Keywords:

  • hypothesis test;
  • confidence intervals;
  • type I and type II errors;
  • one- and two-tailed tests;
  • one and two samples;
  • test of difference between means and population proportions;
  • large and small independent samples;
  • matched samples;
  • differences between variances;
  • equal and unequal variances;
  • assumptions of normality