The Data

The General Household Survey (GHS) is an annual sample survey of households in Great Britain. It collects information on a range of topics from people including smoking, drinking, marriage and cohabitation, housing and consumer durables, and economic activity. The data that you have been given for this assignment is a subset of adults aged 16 years and older from the GHS for 2005. The dataset has 9,590 cases. If you wish to explore the full dataset, it is possible to access the original data from the data archive at www.data-archive.ac.uk. You would need an Athens password to do this (you can obtain an Athens password by visiting the ISS Subscribe website).

There are only seven variables in the reduced dataset you have been given:

AGE – the age of the respondent at their last birthday.

SEX – the sex of the respondent.

NEWSC – new social class

BEN2Q1 – receipt of benefits

SMOKEVER – ever smoked?

CIGNOW – smoke nowadays?

QTYWKDAY – Number of cigarettes a day on weekdays

For Questions 1 and 2 use the GHS data provided.

Question 1

1. What proportion of female respondents have ever smoked? What is the 95% confidence interval for the true proportion of women aged 16+ who have ever smoked in Britain?
   [3]
2. Among those people who smoke nowadays, what is the average number of cigarettes that a person smokes during weekdays? What is the 99% confidence interval for the true mean number of cigarettes smoked per weekday for those who smoke nowadays?
   [3]
3. Are there any differences in the mean number of cigarettes smoked per weekday between young men and women smokers aged less than 30 years? Test at a 5% level of significance. Remember to state your hypotheses and conclusion clearly.
   [4]
4. Is there an association between current smoking status and the social class of a person? Carry out an appropriate statistical test to answer this question, stating your null and alternative hypotheses clearly. If an association exists, what is the nature of this association?

[5]

Question 2

1. Calculate a 99% confidence interval for the true proportion of people aged 16+ in Britain who are in receipt of Income Support. Interpret the confidence interval.
   [3]
2. Recode the variable AGE to have six categories (< 21 years; 21-29; 30-39; 40-49; 50-59; 60+) making sure that the recoded variable is clearly labelled. Produce a frequency tabulation of your recoded variable and state what proportion of the respondents are aged 60+?
   [4]
3. What is the percentage of people aged between 21 and 29 years who are on income support in Britain? What is the percentage of people aged 60+ years in Britain who are receiving pension credit?
   [3]
4. Use the recoded AGE variable to comment on the association (if any) between age and current smoking status. Carry out a Chi-square test and comment on the findings.

[5]

Question 3

You have been asked to carry out a quantitative survey on alcohol consumption and academic performance among students at the University of Southampton. Describe how you might carry out the survey, stating clearly

1. how you would select your sample and why you would choose that method;
2. what sampling frame you would use, commenting on the accuracy of the frame;
3. the type of information on alcohol consumption you would gather and how accurate that information might be;
4. the information on academic performance that you would collect and its accuracy;
5. your views on whether the study would give some indication on the causes of poor academic performance.

Price: $24.01

Solution: The downloadable solution consists of 13 pages, 1101 words.
Deliverable: Word Document