1 Introduct
1 Introduct
What is Statistics?
Areas of statistics
27/04/2012
Diabetes
Cardiovascular
The infant mortality rate of a country is plotted against calendar year for the decade 2000-2008, to see to what extent it is falling.
27/04/2012
A survey of the prevalence of scabies is conducted among school children in two districts. The prevalence is 40% in one district and 20% in the other, but these estimates are based on small numbers of children. The question arises as to whether these data provide good evidence of a difference in prevalence between the two districts, or whether the samples are too small to be conclusive.
District
Data are available on microfilarial counts and nutritional status in adults with onchocerciasis living in a defined area. The data are examined to see whether there is any relationship between microfilarial counts and nutritional status, and in particular whether high microfilarial counts are associated with low weight-for-height.
27/04/2012
A case-control study of hepatoma is carried out to determine whether carriage of hepatitis B is more common among hepatoma cases than among matched controls. The objective is to assess whether carriage of hepatitis B is a risk factor for hepatoma.
Sample size calculation A trial of measles vaccine given by nasal spray is to be conducted in children aged 6 months attending a group or health centre. Statistical techniques are to be used to estimate how many children need to be included in the trial to provide an accurate estimate of the seroconversion rate.
27/04/2012
Types of variables
Variable is any type of observation made on an individual
27/04/2012
Scale of Measurement
Nominal scales
A nominal scale is used when the observation refer to unordered or a classification variables. It simply indicates the group to which each subject belongs.
Ordinal scales
An ordinal scale is one in which the classes do represent an ordered series or relationships. The relationships are expressed in terms of the algebra of inequalities for example a is less than b (a<b) or a is greater than b (a>b).
27/04/2012
Dependent/independent variables
A response variable is an outcome variable whose The explanatory variable predicts or explains other
variables. In an experiment, the explanatory variable is the variable that is manipulated. It is usually a discrete variable. The outcome of the manipulation is determined by measuring the dependent variable. score may be predicted by or related to the independent variable(s). In an experiment, the independent variable is generally discrete and is manipulated while the dependent variable can be discrete or continuous and is measured.
27/04/2012
Variable Quantitative
Qualitative(Categorical)
Rounding
27/04/2012
Class exercise 1
For the following decide type of the variable. Give a range of likely values or a list of categories. Name the individual in each case (the individual may or may not be human) as in the following example. The number of hours of operation of a number of 100 W light-bulbs. individual light bulb type of variable quantitative discrete range of values 0, 1, 2,3,4, 5....100 1. The colour of peoples hair 2. The type of occupation of adult males. 3. The annual rainfall in a country by region. 4.The reaction time of patients to stimulus 5. The position of an individual in the family 6. The grades obtained by candidates taking A-level mathematics. 7. The blood pressure of males 40 and above 8. The number of bacteria in a petri dish. 9. The physical activity of men with high blood pressure. 10. sex of people 11. nationality of employee working in a hospital 12. the presence or absence of a disease in a group of individuals 13. Weight for age (measure of underweight)
In the following examples determine the outcome, explanatory variable, and their types. Give a range of likely values or a list of categories. 1. A double blind trial was conducted to compare the effect of a new treatment with the existing drug (methyldopa) in patients with mild to moderate hypertension. Specifically the study aimed at investigating whether there is any difference in the effects of these drugs on systolic blood pressure. 2. In a survey, the diets of 175 pregnant women were classed as either very poor, poor, fair, good or excellent, according to their nutrient and energy content. The incidence of preeclamtic toxaemia (classified as rise in blood pressure with whether albuminuria or oedema, and excessive weight gain) was also recorded. The aim of the study was to see whether the occurrence of toxaemia is related to the diet in pregnancy.