Statistical Hypothesis

Read Complete Research Material

STATISTICAL HYPOTHESIS

Statistical Hypothesis



Statistical Hypothesis

The data set

The data set is based on a survey of 116 'young' people that smoke and considers variables related to their intention to stop smoking. There are 28 variables with information about demographic factors, ratings on visual analogues scales (perceived risk, self efficacy, subjective norms and intentions to stop smoking), information on previous smoking related illnesses, health value ratings and 18 items from a reliable and valid questionnaire.

The results have been divided into three age-groups namely, 18-21, 22-24 and 25-40. This report consists of four sections in which different hypothesis were tested. The hypotheses are as follows:

•Their will be differences on your scale measures and the MHLC scores according whether they have had a smoking related illness or not.

•There are associations between gender and living accommodation, gender and smoking related illness and accommodation type and smoking related illness.

•There will be an effect of both age and gender on some of your measures.

=> T test for "smoking related illness ever" differences in "age"

One-Sample Statistics

N

Mean

Std. Deviation

Std. Error Mean

Smoking related illness ever

116

.49

.502

.047

Years

116

24.21

4.605

.428

One-Sample Test

Test Value = 0

t

df

Sig. (2-tailed)

Mean Difference

95% Confidence Interval of the Difference

Lower

Upper

Smoking related illness ever

10.540

115

.000

.491

.40

.58

Years

56.612

115

.000

24.207

23.36

25.05

Mean and st.dev of smoking related illness is mean=.49 and st.dev=.502. we run a t test for examine the association among the two variables (smoking related illness and age group) value of t calculated is = 10.540 and 56.612 df = 115 and 115 simultaneousely.

As p value is less than .05 so we can say that smoking related illness ever have a significant differences in age.

Most of the people participate in smoking related illness ever are from age group 22-24, ratio of male participant is relatively high.

=> "smoking related illness ever" in "no of cigarettes"

One-Sample Statistics

N

Mean

Std. Deviation

Std. Error Mean

Smoking related illness ever

116

.49

.502

.047

Reported number of cigarettes before midday

116

2.63

3.412

.317

One-Sample Test

Test Value = 0

t

df

Sig. (2-tailed)

Mean Difference

95% Confidence Interval of the Difference

Lower

Upper

Smoking related illness ever

10.540

115

.000

.491

.40

.58

Reported number of cigarettes before midday

8.299

115

.000

2.629

2.00

3.26

Mean and st.dev of smoking related illness is mean=.49 and st.dev=.502 and for no. of cigrattes mean= 2.63 and st.dev=3.412. we run a t test for examine the association among the two variables (smoking related illness and no. of cigrattes) value of t calculated is = 10.540 and 8.299 df = 115 and 115 simultaneousely.

As p value is less than .05 so we can say that smoking related illness ever have a significant differences in no of cigarettes or we can say that these two variables have effect on each other.

Most of the people participate in smoking related illness ever in no of cigarettes are from age group 22-24, ratio of male participant is relatively high.

=> health value ratings

Independent Samples Test

Levene's Test for Equality of Variances

t-test for Equality of Means

F

Sig.

t

df

Sig. (2-tailed)

Mean Difference

Std. Error Difference

95% Confidence Interval of the Difference

Lower

Upper

Health value

Equal variances assumed

2.692

.104

.039

114

.969

.034

.875

-1.698

1.767

Equal variances not assumed

.039

111.833

.969

.034

.875

-1.698

1.767

For Health value we run independent sample t test value of F calculated in above table is 2.692 df=114 95% C.I is -1.698 and 1.767.

=> smoking related illness ever differences in visual analogue scale

Smoking related illness ever

Frequency

Percent

Valid Percent

Cumulative ...
Related Ads