Statistical Significance: A Biostatistics Approach

Test of Statistical Significance

The mathematical tool used to compare data between 2 or more groups.

Parametric Tests

• Used when quantitative variables are used.
• Compare means and standard deviations
• Normal distributions
• More powerful

Non-parametric tests

• Used when qualitative variables have been used.
• Compare percentages and proportions
• Skewed distributions
• Less powerful

In biostatistics, there are 2 types of variables:

• Quantitative variables: These cannot be measured or compared.
• Qualitative variables: These can be measured and compared. For comparison, they have assorted measuring units.

Types of Parametric Tests

Understand Test of Statistical Significance with the help of following Flow Charts :

Frequently Asked Questions  

Q1. If we have to compare mean hemoglobin in a group of 20 individuals before giving iron supplements and after giving iron supplements, which test will be used?

Ans: As we are comparing only one group of 20 individuals, it is an example of a paired T-test.

Q2. We have to compare mean hemoglobin in a group of malaria patients before and after treatment.

Ans: Paired T test is used in this comparison.

Q3. Mean hemoglobin in 20 individuals. Out of these, 10 individuals were given IFA supplements, and another 10 individuals were given dietary supplements.

Ans: As we are comparing the mean hemoglobin in two groups, the Unpaired T-Test is used.

Q4. Compare mean hemoglobin in a group of malaria patients and dengue patients

Ans: Unpaired T-test is used since 2 groups are compared

Q5. Compare mean hemoglobin in a group of malaria patients, dengue patients, and Chikungunya patients.

Ans. ANOVA is used as three groups are compared.

Q6. On campaigning the mean blood pressure in 3 communities:

A. Community 1 is on dietary supplements

B. Community 2 is on medical treatment

C. Community 3 is on exercise

Ans. ANOVA is used as three groups are compared.

Q7. Which statistical test are you going to use when we have to study the association between OCP use and breast cancer?

Ans. As breast cancer is a qualitative variable, a non-parametric test is used . Hence, the Chi-Square Test is used.

Q8. Compare the prevalence of obesity in a group of 20 women before a dietary regime and after the dietary regime.

Ans. As it is for one group, Wilcoxon Signed Rank Test or McNemar test is used. However, most commonly Wilcoxon signed rank test is used.

Q4. Compare mean blood pressure in a group of 20 women before and after a dietary regime.

Ans. In a single group, a Paired T-Test.

• Quantitative variables: Mean and SD
• Qualitative variables: Percentage and Proportions
• 1 group: Paired Test
• 2 groups: unpaired test
• More than 2 groups: ANOVA

Important Exam related MCQs based on Statistical Significance

Q. All of the following are non-parametric tests except?

1. Chi-square test
2. Z test
3. Wilcoxon rank sum test
4. Kruskal Wallis H test

Answer B: Z test

Q. A study to compare the hemoglobin level was conducted on alcoholics before and after consumption. The statistical method used to find significance Is?

1. Chi-square test
2. Unpaired T test
3. Paired T-test
4. Mann Whitney test

Answer C: Paired T-test

Q. The test used to compare two qualitative data is?

1. Paired T-test
2. Unpaired T test
3. Chi-square test
4. Anova

Answer C: Chi-square test

Q. All of the following can be analyzed using the chi-square test except?

1. Sex and stage of cancer
2. Heart rate/min and age
3. Benign or malignant, and Type of surgery
4. Age group and cancer stage

Answer B: Heart rate/min and age

Q. The appropriate statistical test to find out if obesity is a significant risk factor for breast cancer is?

1. Students paired T-test
2. Students unpaired T-test
3. Chi-square test
4. Wilcoxon signed-rank test

Answer C. Chi-square test

Q. Three groups of subjects were followed over the course of five years to compare treatments for sideroblastic anemia. The most appropriate statistical analysis to determine the quantitative serologic differences resulting from these treatments would be

1. Regression analysis
2. ANOVA
3. Correlation analysis
4. Chi-square test
5. T-test

Answer : B. ANOVA

Q. The weight of the group of 50 boys aged 12 years was 35+5 kg & 40 girls the same age were 32 kg+3 kg. Test applicable to test statistical significance difference in weight

1. Chi-square test
2. Z test
3. T
4. ANOVA

Answer: B. Z test

Download the PrepLadder app now and unlock a 24-hour FREE trial of premium high-yield content. Access Smarter Video Lectures also in हिंglish, Game Changing Qbank, Audio QBank, Structured Notes, Treasures, Mock test for FREE to ace your NEET PG preparation. Elevate your study experience and gear up for success. Start your journey with PrepLadder today!