Ari Julianto says:


You may copy and share the contents in my blog, but please cite my blog address as your reference. I only accept the comments that relate to the postings in this blog. For private and personal issues, you may contact me at
mr.ari69@gmail.com

Two Types of Hypothesis


                    Two Types of Hypothesis
 



 Written by John W. Creswell in Educational Research, Planning, Conducting, and Evaluating Quantitative and Qualitative Research. Boston: Pearson.2012.pp. 126-129.


There are two types of hypotheses: the null and the alternative to the null. You need both types in a research study, but authors generally write only one or the other into their reports.


1. Null Hypotheses
The null hypothesis is the most traditional form of writing a hypothesis. Null hypotheses make predictions that of all possible people whom researchers might study (i.e., called the general population), there is no relationship between independent and dependent variables or no difference between groups of an independent variable or a dependent variable.

To study this hypothesis, you would select a sample of all possible people and draw conclusions from the statistical analysis of this sample for the population. A null hypothesis might begin with the phrase “There is no difference between” groups or “There is no relationship between (or among)” variables.

To write a hypothesis, you can complete the following script, which employs the language “no difference”:
There is no difference between (independent variable, group 1 ) and (independent variable, group 2) in terms of (dependent variable) for (participants) at (research site).
An example of the application of this script might be:

There is no difference between at-risk and non-at-risk  students in terms of student achievement on math test scores for third-grade students in a Midwest school district.
Independent variable: at-risk students (members and nonmembers) Dependent variable: student achievement test scores Participants: third-grade students
Site: X school district
Form and language: null indicating no difference

2. Alternative Hypotheses
In contrast to the null hypothesis, you may write an alternative hypothesis. You will use an alternative hypothesis if you think there will be a difference based on results from past research or an explanation or theory reported in the literature.

The two types of alternative hypotheses are directional and nondirectional. In a directional alternative hypothesis, the researcher predicts the direction of a change, a difference, or a relationship for variables in the total population of  people. A researcher selects a sample of people from a population and predicts that the scores will be higher,better, or changed in some way. This typical form for writing hypotheses is encountered in the literature more than any other type of hypothesis.

A script for a directional alternative hypothesis is:
(group 1, independent variable) at (research site) will have (some difference, such as higher, lower, greater, lesser) on (dependent variable) than (group 2 of independent variable).
An example of this script is:

Students who participate in direct learning in four elementary schools will have higher achievement scores than students who participate in whole-language learning.
Independent variable: learning (direct and whole language)
Dependent variable: achievement test scores
Participants: third-grade students
Research site: four elementary schools
Key indicator: directional, a prediction is implied alternative hypothesis the researcher predicts a change, a difference, or a relationship for variables in a population but does not indicate whether the direction of
this prediction will be positive or negative, or greater or less.

The nondirectional alternative is not as popular as the directional alternative because the researcher does not take a stand about the direction of the relationship of the variables. A script for a nondirectional alternative hypothesis is:
There is a difference between (group 1, independent variable) and (group 2, independent variable) in terms of (dependent variable).

An illustration of this script would be:
There is a difference between varsity athletes in high school who smoke and those who do not smoke in terms of athletic accomplishments.In this example, the author does not state whether the difference will be positive or negative. An analysis of the variables in this statement shows:
Independent variable: use of tobacco (smokers and nonsmokers)
Dependent variable: athletic accomplishments
Participants: varsity athletes
Sites: high schools
Key indicator: the words “a difference,” but the direction is not specified.


Related Posts Plugin for WordPress, Blogger...