Question

I need to know what kind of t-test to use for this problem, and how to figure it out in SPSS ONLY.

SCENARIO 1: Previous research indicates that children borne by diabetic mothers may suffer from high blood pressure, among other health problems. A group of researchers collected data on the systolic blood pressures, in mm Hg, of 99 participants borne by diabetic mothers (diabet). Systolic blood pressure is considered within a normal range if it is under 120 mm Hg. At the 1% significance level, test the hypothesis that the systolic blood pressure of children borne by diabetic mothers is different than 120 mm Hg.

[To test this hypothesis, you will need the variable labeled ‘diabet’].

Answer 1

Tests of significance

There are several parametric & nonparametric test available to establish the statistical significance.

Parametric test

These are also known as normal distribution statistical tests. The stastistati methods of inference make certain assumptions about the populations from which the samples are drawn.

For example, the assumptions may be that the population are normally distributed, have the same variance etc. ; population values, as we have seen, are known as parameters; the statistical techniques which make assumptions about the parameters are csllca parametric techniques.

t-test

Z-test

ANOVA

Nonparametric test

Chi-square test

Median test

Mc Nemar test

Mann-Whitney Test

Wilcoxon test

Fishers exact test

t - Test

It is applied to find the significant difference between two means. This test can be applied when following criteria are fulfilled

• Randomly selected homogeneous sample. ( Here the samples is homogeneous ie only children of diabetic motheri are selected )
• Qualitative data ( Here the data of systolic blood pressure is provided in mmHg )
• Variability normally distributed
• Sample size less than 30, if the sample size is more than 30, then Z test is applied. However, even t- test can be applied in sample size more than 30. ( Here the sample size is 99 )

This study fulfilled all the desired criteria for t-test

Types of t-test

Unpaired t-test

It is applied if we obtain data from subjects of two independent separated groups of people or samples drawn from two different population ( In this scenario research collected data from the the same population or children of diabetic mother & only one group is present here. So we can't use this type here )

Paired t-test

It is applied on paired data of independent observations made on same sample before & after the intervention. Paired test is most commonly used in Nursing Research Studies. ( Here all 99 sample is observed independently by the researcher & the sample is undergoing a second study. )

So Paired t - test is the most appropriate choice in this study