SPSS- Data Analysis Essay Example

• Category:
Other
• Document type:
Math Problem
• Level:
• Page:
2
• Words:
948

TASK 6: Examination of the relationship between gender and height

Generalisation indicates that on average males are taller than men, even though indidual women may be taller than some men. Inorder to investigate this generalization a t-test was used.

1. The two variables to be used were gender and height in metres where gender is the independent variable while height is the dependant variable. Gender is qualitative and is classified as nominal, while height is quantitative and is classified as ratio

1. The one tailed independent t-test was found to the most appropriate as there are several reasons that are favourable to this test because it involves comparison of a quantitative variable (age) in two groups ie a quantitative variable are involved and thus the test is suitable. Paired t-test was not considered because of lack of dependance between the

1. H0: the mean height for men is either equal or less than that of female

H1: the mean height of men is grater than mean heat for female

H0: µmale ≤ µfemale H
A: µmale> µfemale

1. This test required the manupulation of data before it was done because because the height was to be found by relatinf BMI and the mass (weight). Having transferred both the BMI variable and the weight variable into SPSS, the compute function was used in creating the new height variable.

1. The assumption in performing the t-test was that the subjects were randomly selected, there was independence of the subjects, data normally distributed and homogeinety of varience exhibited. Random selection of subjects and independence of the subjects was part of put into consideration at research design stage. Normalility was tested through Q-Q plots of height. Homogeineity of variance was tested by use of Levene,s test where for p<α where α=0.05 the assumption of homogeinety will have been violated. In this case Levene’s statistics was 0.171 and thus homogeneity of variance exhibited (table 2)

1. The t-test resut were as shown in table 1 and table 2. From table 1 it can be seen that the mean height of males was higher at 1.77 with that of female being lower at 1.64

Table 2 shows that the mean diffrence was statistically significant t(250)= -15.577 (p<0.01).

1. This test has proved that our assumption that the mean height of males is statistically significant from that of women which proves that the general assumption that in anygeneral population men will be found to be taller. The fact that the test was statistically significant at p<0 means that there there is only a probability of 1% that there may ot be true and was only as a result of sampling error.

1. If the data for 2009 was used there could be a slight diffrence in average heights since beyond 20 years of age research has shown that there is a reduction in height (Sorkin J. D. 1999). In order to avoid this shortcoming it would be recommended that a norrow age range be used like in 30s so as to have a group with similar age changes.

1. The Australian Bureau of statistics National Health Survey of 1995 has indicated that the average height of males was 174.8cm while that of female was 161.4cm. this data shows that the sample means of this study was greater than that of 1995 survey with males having 2.25cm high value and females having 3.02cm higher value.

6.42×10-3

TASK7: Examination of the relationship between diet and physical activity

1. two variables that are being used are diet and physical activity where the former is independe and the later is dependant. both the two variables are qualitatitive and they can be classified as nominal.

2. With the two variables being nominal chi-square is used in testing the hypothesis as this is the most versatile test. The test will be used in finding out if there is any relationship between the two variables.

1. HO: There is no relationship between diet and physical activity

H1 : There is a relationship between diet and physical activity

( Here no symbols are used thus no notation) (SSB, 2011).

1. The data did not need any form of manipulation apart from coding appropriately.

1. Chi-square did not require any special conditions to be mate due to its versatility.

1. The test result were as shown in table 3 and table 4. From table 3 it can be seen that a bigger proportion of those who have low activity consume a low diet food where 128 out of the 185 low activity consumed low diet food. For the group with high physical activity out of a total of 96 in the group 37 took a high diet food which is above the expected 25.3. this result is reflected in the table 4 that gives the chi-squre result. From the table it can be seen that there is a significant relationship between the variables 11.758 (p<0.01). this results are diagramatically represented as shown in figure 2.

1. To find out if gender had no effect on the relationship a cross tabulation was done with gender being incorporated as a layer and the chi-square result was as shown in table 5. From the results it can be observed that there was a significant relation between the two variables (diet and physical ability).

1. From this analysis it has been seen that the low activity respondents were associated with low diet food while high activity was associated with high quality diet. This is in agreement with what has been found in other studies.

References

Stident services Building (SSB).(2011). Statistics: The Null and Alternate Hypotheses

A Student Academic Learning Services Guide. Durhum college

Sorkin J. D. (1999). Longitudinal Change in Height of Men and Women: Implications for

Interpretation of the Body Mass Index. American Journal of Epidemiology