HYPOTHESIS TESTING

Hypothesis Testing

Hypothesis Testing: If a Company Provides an On-Site Daycare, Then Employees Will Miss Less Work

Introduction

The primary means of conveying the strength of empirical findings in social and management science is the hypothesis significance test. This paradigm, along with its strengths and weaknesses, is therefore important for nearly every quantitative study in social and management science. This paper will reviews the current hypothesis testing paradigm and will present a practical application of the hypothesis testing using some real fact based data. The paper will also discuss the underused idea of statistical power from tests, and points out some common misinterpretations of hypothesis testing. It is important to examine the nature of the hypothesis before the determination of the test. Hypothesis can be divided into two types (Ary, Jacobs, Razavieh & Sorensen, 2006):

Directional, and

Non-Directional

Directional Hypothesis

A directional hypothesis is a prediction made by a researcher regarding a positive or negative change, relationship, or difference between two variables of a population. This prediction is typically based on past research, accepted theory, extensive experience, or literature on the topic. Key words that distinguish a directional hypothesis are: higher, lower, more, less, increase, decrease, positive, and negative. A researcher typically develops a directional hypothesis from research questions and uses statistical methods to check the validity of the hypothesis.

Examples of Directional Hypotheses

A general format of a directional hypothesis would be the following: For (Population A), (Independent Variable 1) will be higher than (Independent Variable 2) in terms of (Dependent Variable). For example, “For ninth graders in Central High School, test scores of Group 1 will be higher than test scores of Group 2 in terms of Group 1 receiving a specified treatment.” The following are other examples of directional hypotheses (Moore & McMabe, 1993):

There is a positive relationship between the number of books read by children and the children's scores on a reading test.

Teenagers who attend tutoring sessions will make higher achievement test scores than comparable teenagers who do not attend tutoring sessions.

Nondirectional and Null Hypotheses

In order to fully understand a directional hypothesis, there must also be a clear understanding of a nondirectional hypothesis and null hypothesis.

Nondirectional Hypothesis

A nondirectional hypothesis differs from a directional hypothesis in that it predicts a change, relationship, or difference between two variables but does not specifically designate the change, relationship, or difference as being positive or negative. Another difference is the type of statistical test that is used. An example of a nondirectional hypothesis would be the following: For (Population A), there will be a difference between (Independent Variable 1) and (Independent Variable 2) in terms of (Dependent Variable 1). The following are other examples of nondirectional hypotheses (Moore & McMabe, 1993):

There is a relationship between the number of books read by children and the children's scores on a reading test.

Teenagers who attend tutoring sessions will have achievement test scores that are significantly different from the scores of comparable teenagers who do not attend tutoring sessions.

Null Hypothesis

Statistical tests are not designed to test a directional hypothesis or nondirectional hypothesis, ...