Exercise 1: What is the correlation

between smoking on week days and smoking on weekends among older people?

To be able answer this question, the researcher has to

filter out the dataset to include respondents who are 60years and above. The

total number of respondents irrespective of the age level were 10,601 and after

it was filtered to include those who were 60years and above, the number stood

at 7664. The number of missing observation for HeSkb is 7,109 and that of the HeSkc also 7,109.

Table 1.1: Correlations between HeSkb and HeSkc

Number of cigarettes smoke per weekday

Number of cigarettes smoke per weekend day

Number of cigarettes smoke per weekday

Pearson Correlation

1

0.934

Significance value

0.000

Total Number

555

555

Number of cigarettes smoke per weekend day

Pearson Correlation

0.934

1

Significance value

0.000

Total Number

555

555

Source:

Researcher’s Own Calculation, 2018

From the correlation analysis table as indicated in

the Table 1 above, the association between the variables was approximately 93%

which indicates high level of the strength of the association and this

association is being confirmed by the small p-value of 0.000 at 5% significance

level, which indicates high level of significance between the two variables.

This means that number of cigarettes smoke per weekdays is highly correlated

with number of cigarettes smoke per weekend.

1 (a) Figure 1.1:

Plot of Heskc against Heskb

Source:

Researcher’s Own Calculation, 2018

1

(b) Figure 1 above shows the scatter

plot of the HeSkb against HeSkc. From the figure, it can be observed that there

is an indication that there is a strong and positive relationship existing

between the two variables understudy.

1 (c)

Source: Researcher’s Own Calculation, 2018

Exercise 2: What are the Effects of

Care Provision, Age and Nature of Reciprocity of Life Satisfaction among Older

People who provide Care?

Table 2.1

Statistics

Sex

age

Hours

spent looking after other people last week

Respondent

is satisfied with what they have gained so far from caring for others

Respondents feel they have been

adequately appreciated for caring for others

In

most way, his/her life is close to his/her ideal

The

conditions of his/her life are excellent

Is

satisfied with his/her life

So

far, he/she has gotten the important things wants in life

If

could live his/her life again, would change almost nothing

Valid

10601

10601

1935

2725

2722

8737

8713

8838

8807

8816

Missing

0

0

8666

7876

7879

1864

1888

1763

1794

1785

Source: Researcher’s Own

Calculation, 2018

Source:

Researcher’s Own Calculation, 2018

According to (William Pavot & Ed Diener, 2008),

they indicated that SWL values range from 5-35. They stated that SWl value of 20

indicates a neutral point when using the SWL scale. The study indicated that

values between 5-9 means that the respondents are extremely dissatisfied in

their way of life. Whiles those with scores between 31-35 represent those who

are extremely satisfied with their way of life. Values between 21-25 years were

considered slightly satisfied and 15-19 indicating slightly dissatisfied in

life.

Table 2.2: Sum All

Frequency

Percentage (%)

Percentage (%)

Neutral

306

2.9

3.4

Extremely

dissatisfied

280

2.6

3.1

Extremely

satisfied

1418

13.4

15.9

Slightly

satisfied

1749

16.5

19.6

Slightly

dissatisfied

927

9.2

10.9

Satisfied

580

5.5

6.5

Extremely

satisfied

3607

34.0

40.5

Total

8912

84.1

100.0

System

1689

15.9

–

Total

10601

100.0

–

Source:

Researcher’s Own Calculation, 2018

RECODE sum_all (20=1) (5 thru 9=2) (31 thru 35=3) (21 thru 25=4)

(15 thru 19=5) INTO sumall.

EXECUTE.

RECODE sum_all (20=1) (5 thru 9=2) (31 thru 35=3) (21 thru 25=4)

(15 thru 19=5) (10 thru 14=6) (26 thru 30=7) INTO sumall.

EXECUTE.

FREQUENCIES VARIABLES=sumall

/ORDER=ANALYSIS

Source:

Researcher’s Calculations, 2018

d. Create two new dummy variables

measuring the reciprocal relationships in care giving by recoding ErCarA and

ErCarB: Recode 1 and 2 to 1, 3 and 4 to 2 so that 1 indicates “strongly

agree/agree” and 2 indicates “disagree/strongly disagree”.

After Recoding

Table 2.3 (Ner)

Frequency

Percentage

(%)

Valid

Percentage (%)

Refusal

4

0.0

0.0

Item

not appropriate

7838

73.9

74.2

Strongly

agree/agree

2528

24.6

23.9

Disagree/strongly

disagree

118

1.1

1.8

Total

10567

99.7

100.0

System

34

0.3

–

Total

10601

100.0

–

Source:

Researcher’s Calculations, 2018

Table 2.4 (Nerb)

Frequency

Percentage

(%)

Valid

Percentage (%)

Refusal

5

0.0

0.0

Item

not appropriate

7838

73.9

74.2

Strongly

agree/agree

2528

23.8

24.7

Disagree/strongly

disagree

194

1.8

1.1

Total

10567

99.7

100.0

System

36

0.3

–

Total

10601

100.0

–

Source:

Researcher’s Calculations, 2018

2.2 (a)

i) The appropriate regression method to

fit the model 1 is the Simple Linear regression. This method fit the data well

because it uses one dependent and one independent for the analysis.

(ii).The regression method that fit

the second model 2 is the Multiple Regression technique. The model is appropriate

because it uses one dependent and more than two independent variables.

(b)

Table 2.5: Coefficients for the Two Models (Simple Linear

and Multiple Linear Regression)

Model

Unstandardized

Coefficient

Standard

error

Standard

coefficients

t-ratio

Significance

value

B

B

Simple

linear regression

Constant

24.946

0.074

336.024

0.000

Hours

spent looking after other people last week

-0.015

0.002

-0.068

-6.459

0.000

Multiple

linear regression

Constant

32.533

1.418

22.948

0.000

Hours

spent looking after other people last week

-0.017

0.002

-0.136

-6.896

0.000

Dum1

-2.919

0.691

-0.086

-4.227

0.000

Dum2

-3.768

0.534

-0.144

-7.057

0.000

Sex

-0.054

0.275

-0.004

-0.197

0.844

Age

0.000

0.015

-0.001

-0.030

0.976

Source:

Researcher’s Calculations, 2018

Coefficient of Determination Table for the Two Models

(Simple and Multiple Linear Regression)

Regression Model

R

R-Square

Simple

Linear Regression Model

0.068

0.005

Multiple

Linear Regression Model

0.235

0.055

Source: Researcher’s Calculations, 2018

(c)

2.3 (a)

Table 2.6: Coefficients

for the Two Model (Simple and Multiple Linear Regression)

Model

Unstandardized

Coefficient

Standard

error

Standard

coefficients

t-ratio

Significance

value

B

B

Multiple

linear regression

Constant

32.533

1.418

22.948

0.000

Hours

spent looking after other people last week

-0.017

0.002

-0.136

-6.896

0.000

Dum1

-2.919

0.691

-0.086

-4.227

0.000

Dum2

-3.768

0.534

-0.144

-7.057

0.000

Sex

-0.054

0.275

-0.004

-0.197

0.844

Age

0.000

0.015

-0.001

-0.030

0.976

Source:

Researcher’s Calculation, 2018

Table 2.7: Coefficient of

Determination for the Multiple Linear Regression Model

Regression Model

R

R-Square

Multiple

Linear Regression Model

0.235

0.055

Source:

Researcher’s Calculations, 2018

The result

in the Table 2.7 provides the coefficient statistics for the variables under

consideration. From the result as indicated in the table, hours spent looking

after other people last week (ErCAC) is statistically significance having

impact on the Satisfaction with life.

Also, the dummy variables created by the researcher were

all statistically significance at 0.05. The dum1 and dum2 have small

significant p-values of 0.000, which are less than 0.05 alpha level.

Furthermore, sex of respondents was not significant at

0.05. Its means that sex does not have impact on the SWL.

Finally, age of respondents is not significant at

0.05. It means that the ages of the respondents have no impact on the

satisfaction level in the lives of the respondents.

2.3 (b)

Life satisfaction is what every

individual is expecting to have. According to a study done by (Deary, Corley, Gow,

et al, 2009), they were of the view that ageing is usually associated with

declining economic resources, decreasing cognitive ability, deteriorating

physical health and weakening social support especially among older people in

society. This means that in most case, the satisfaction level among the older

people decline. The study conducted by the researchers titled “what Matters for

Life Satisfaction among the Oldest-Old?”

indicated that when it comes to life satisfaction, more women rated

themselves good or very good to enjoy life satisfactory as compared to the men.

The result obtained by the women is giving as (?=-0.308, 95% CI = -0.438 to -0.177,

p