INTRODUCTION TO SPSS (Hypothesis Testing)

In Statistical Analysis, we can only conclude or generalize our result unto the population for decision making process. The process of achieving this is known as hypothesis testing.

In this article, I shall elaborate intensively on hypothesis, level of significance, null and alternate hypothesis and steps to carryout hypothesis testing.

Kindly read through carefully this guide on hypothesis testing in SPSS.

What is Hypothesis?

Hypothesis is a statement that aids to generalize our test result to the entire population when compared with a value. This is also known as statistical power of that test. It helps to check if our analysis is valid or not.

What is Hypothesis Testing?
Hypothesis Testing is an inferential statistics test in SPSS, it is the logic of drawing conclusion generally to the population from samples of data collected.

For a valid analysis, we can believe or trust the result hence we can generalize it to the entire population. For example I want to carry out a research if Ford cars consumes fuel a lot.

I have to pick samples of Ford cars in my locality in order to be able to generalize or prove the statement concerning the entire Ford cars in the world. This process is known as hypothesis testing.

To aid hypothesis testing, all test carried out using SPSS software usually has its own hypothesis and a value of significance usually displayed in the output window.

Types of Hypothesis
There are two types of Hypothesis namely;

Null hypothesis
Alternative hypothesis

Null hypothesis
A null hypothesis is a statement about the sample which may be true or false concerning the entire population.

Do you still remember what we refer to as a population? If no read the heading on hypothesis testing again before continuation.

In null hypothesis, the hypothesis is either accepted or rejected. It is denoted by the letter Ho.
If we want to test significance of difference by two sampled statistics, we adopt the null hypothesis.

E.g if we want to find out if eating a particular food increases weight, we assume both samples from the population are not different. That is the null hypothesis otherwise known as Ho.

The test model we will use is "test of difference" and the null hypothesis for test of difference states that there is no difference between both samples.

It is either you accept or reject the null hypothesis based on a significanct value usually .05 meaning that five samples out of 100 rejects the null hypothesis.

Don't get scared of the value as it shall be elaborated on as we proceed further.

I hope we are on the track together? If true then you may proceed to the next paragraph.

Alternative hypothesis
Alternative hypothesis is the opposite of a null hypothesis. It is denoted by H1.
That is to say if the null hypothesis is not true, we choose the alternative hypothesis.

It is done base on significance level test.

However, there is usually a problem of wrongly rejecting a true null hypothesis and accepting a false null hypothesis. This posses a problem known as type I and type II errors.

Errors in Hypothesis Testing
The following errors are discovered in hypothesis testing.

Rejecting a null hypothesis when it is true otherwise known as type I error.

Accepting a null hypothesis when it is false otherwise known as type II error.

In business management, they are also known as producer and consumer risk (Gupta, 2021).

Type II error is of more consequence than type I.

We make type I error by rejecting a true null hypothesis.

We make type II error by accepting a wrong null hypothesis.

Remember in a null hypothesis, you either accept or reject the statement .

Mathematically we can say:

The possibility of rejecting Ho when it is true = P(type I error) = alpha

The possibility of accepting Ho when it is wrong = P(type II error) = beta

Where alpha is the size of type I error.
And beta is the size of type II error.

In other words;

Type I - RT- Rejecting good
Type II - AF - Accepting bad

Where RT implies rejecting true
And AF implies accepting false

We can also say:
Beta = P (type II error) = P (accepting Ho when false)

Alpha = P (type I error) = P (rejecting Ho when true)

P(type II) + P(type I) = 1
The probability of making type I error in addition to type II error equates to 1.

Since we don't want to make the mistake of accepting the bad, it is better accepting when it is true.

Then we say that P (type I) = 1 - P (type II)
alpha = 1 - beta

Maximizing alpha which is the possibility of rejecting true null hypothesis, we need to use a level of significance.

Selecting maximum size of type I error prepared to be risked is referred to as level of significance also known as alpha.

Level of significance choosen by scientists are 5% and 1%

It is understood that whenever we reject a true null hypothesis which is alpha, we are confident in our decision depending on the level of significance .

When the null hypothesis is accepted, it means that data does not provide evidence against the null hypothesis to reject it.

Note that, If the value of P is < alpha level of significance (0.05 , 0.01) you have to reject the null hypothesis

If P value is > alpha , accept the null hypothesis at such level of significance. It means the data doesn't provide sufficient evidence to reject the null hypothesis.

Steps for Checking Hypothesis

Select the proper statistics test for your data.

State the null hypothesis and alternate hypothesis

Select the level of significance

Calculate the statistics

Make the decision

Conclusion
This guide has been able to put you through the first step of analysis in SPSS which is hypothesis testing i.e stating the null hypothesis and alternative hypothesis. After this the next phase is to either accept or reject it for decision making process. If there be questions, suggestions and comments as regards this subject please do well to indicate in the comment section of this post as I will be available to answer, add and view them
This topic helps a lot as it is the backbone of decisions made by the government.