Tutorial 1 STF1103 Subgroup 1 PDF

Preview

Summary

STF 1103 STATISTICS FOR BIOLOGY 2TUTORIAL 1LECTURER: DR SITI NURLYDIA BINTISAZALI @ PIKSINDUE DATE: 28TH MARCH 2022GROUP MEMBER MATRIC NUMBEREMMANUEL ENJAH ANAKDEREK79319ELICIA SHEMA ANAKEDWIN LARKIN79275ZULHILMI BIN MOHAMADISMAIL81569LUKMAN HAKIM BINHAIRUDDIN781161. Discuss what a nonparametric tes...

Description

STF 1103 STATISTICS FOR BIOLOGY 2 TUTORIAL 1 LECTURER: DR SITI NURLYDIA BINTI SAZALI @ PIKSIN DUE DATE: 28TH MARCH 2022

GROUP MEMBER EMMANUEL ENJAH ANAK DEREK ELICIA SHEMA ANAK EDWIN LARKIN ZULHILMI BIN MOHAMAD ISMAIL LUKMAN HAKIM BIN HAIRUDDIN

MATRIC NUMBER 79319 79275 81569 78116

1. Discuss what a nonparametric test is. (4 M) A nonparametric test is a hypothesis test that does not require the population's distribution to be characterized by certain parameters. This test did not have assumption like parametric test usually rely on. Thus, this nonparametric test is useful when our data is non normal and resistant to transformation. However, nonparametric tests do not completely rule out the assumptions about the data. For example, the nonparametric test requires the data to be an independent random sample. Non parametric test is easier to understand and not too complicated compared to parametric test. 2. Describe the advantageous and disadvantageous of a nonparametric test.

(10M)

Advantageous

Disadvantageous

Nonparametric test is simple and easy to understand because it used under more general conditions (robust).

Nonparametric test only can be applied for nominal or ordinal scale data because any count measurement data may be ignored or wasted. The results obtained may be more general when little information or no information available regarding the population. Thus, stronger support must be established before reject a null hypothesis. Not so efficient and less powerful than the parametric test. Thus, it less likely to reject the null hypothesis when it is false.

Computation is usually easy and not complicated because it does not have many assumptions relating to the data format that is common in parametric tests Available for nominal scale data and ordinal scale since the data being distribution-free. This can also be said as less sensitive and less efficient.

No assumption is made regarding the parent population.

Not available for testing the interaction in analysis of variance model (ANOVA).

3. Identify the importance of a nonparametric test in biological studies.

(6 M)

➢ This test is important for data that have a very small sample size for example the sample size of a population is less than 30. ➢ It is an important test for some data that are strongly non normal and resistant to transformation ➢ Appropriate when examining ordinal or nominal data when the assumptions of parametric test have not been achieved.

4. Restate the key statements given and complete the crosswords.

(10 M)

Key statements: 1. The Mann-Whitney rank sum test is a counterpart of the t test to compare the means of two independent populations. 2. The nonparametric test that is analogous to one-way ANOVA is the The Kruskal-Wallis Rank Sum Test. 3. The Wilcoxon signed-rank test is a non-parametric alternative to the t test for related samples. 4. The One Sample Sign test is the nonparametric alternative to the one sample t test and z test. 5. The Friedman test is the nonparametric alternative to the Two-way ANOVA.

6. Nonparametric tests can be applied to nominal and ordinal scaled data. 7. The null hypothesis is rejected when the test statistics calculated is less than the critical value. 8. The median test is used to compare two or more groups that are independent. 9. Parametric tests are based on some restrictive assumptions about the parameters. 10. Nonparametric tests are not based on the restrictive normality assumption of the population or any other specific shape of the population....