Week 12 Data Analysis and Presentation
Week 12 Data Analysis and Presentation
• Interval estimates
– Parameter estimated from a sample data (point estimate or sample estimate)
as opposed to population (true value) parameter.
• Mean – the true mean is the sum of all the members of the given population
divided by the number of members in the population. Impractical to measure every
member a random sample is drawn gives the point estimate of the
population mean.
– Interval estimate expand on point estimates by incorporating the uncertainty
of the point estimate.
• For example, different samples from the same population will generate different
values for the sample mean.
• An interval estimate quantifies this uncertainty in the sample estimate by
computing lower and upper values of an interval which will, with a given level of
confidence (i.e., probability) contain the population parameter.
Measures of central tendency continued…
• Why different measures
– Normal distribution
• Symmetric distribution – single peak, well-behaved tails
(estimates for mean, median & mode similar) - use mean as the
locator estimate.
– Exponential distribution
• Skewed distribution – mean & median not the same – mean pulled
to one side (direction of skewness).
Use all three central measures.
– Cauchy distribution
• Symmetric distribution – single peak with heavy tails
extreme values in the tails distort the mean - use median as the
locator estimate.
Quantitative techniques continued…
• Hypothesis test
– Also addresses the uncertainty of the sample estimate.
However, instead of providing an interval, a hypothesis test
attempts to refute a specific claim about a population
parameter based on the sample data.
• To reject a hypothesis is to conclude that it is false.
• To accept a hypothesis does not mean that it is true, only that we
not have evidence to believe otherwise.
– Hypothesis tests are usually stated in terms of both a
condition that is doubted (null hypothesis) and a condition
that is believed (alternative hypothesis).
Quantitative techniques continued…
• Common format for a hypothesis test:
– H0 : a statement of the null hypothesis, e.g., two population
means are equal.
– Ha: a statement of the alternative hypothesis, e.g., two population
means are not equal.
• Ascription
• Pre- and post-measures of change.
• After-the-fact statements of change
• Explicit statements of cause/motivation of change
• Evidence ruling out plausible alternative explanation for the
change
• Independence evidence attesting to the program’s likelihood of
effecting change.
Data interpretation continued…
• Assessment
• Comparison with past project performance
• Comparison with accepted target levels
• Comparison with other programmes or general norms
• Comparison with constituents needs
• With some standards, cost-benefit comparison
Data interpretation continued…
• Description of the sample
– Describe the study population by producing tables showing the
distribution of important variables e.g. sex, age, sex by age,
morbidity, nutritional status, nutritional status and age, nutritional
status and sex, nutritional status and morbidity, etc.