Analytical Phase
Data analysis
- converting data to information
Branches:
Descriptive Statistics
-methods applied in order to organize, summarize and present data in
a form which will make them easier to analyze and interpret
Inferential Statistics
-methods involved in order to make generalizations and conclusions
about a target population, based on results from a sample
Variables
-a value of a characteristic that changes from subject to subject or
from time to time
Relationship between Variables:
Dependent Variable
- the outcome or response variable
Independent Variable
- the variable that is presumed to influence the outcome
Confounding Variable
- An extraneous variable which by itself may influence the outcome and
which may be mistaken to be the effect of the independent variable
being considered.
Types of Variables:
Qualitative Variables
- categories are simply used as labels to distinguish one group from
another
Quantitative Variables
- categories can be measured and ordered according to quantity/amount
Discrete Quantitative Variables
- can assume only integral values or whole numbers
Continuous Quantitative Variables
- can attain any value including fractions or decimals
Scales of Measurement
Nominal Scale
- unordered categories
Ordinal Scale
- ordered categories
Interval Scale
- the exact distance between two categories can be determined but the
zero point is arbitrary
Ratio Scale
- Similar to interval but the zero point is fixed
Kinds of indicators
- Parameters – derived from the entire population
- Statistics – derived from a sample of the population
Measures of Central Tendency
Mean – arithmetic average
Median – middle value
Mode – most frequently occurring value
Measures of Variation
Range
- the difference between the highest observation and the lowest
observation
Variance
- refers to how much the individual observations differ from the mean
Standard Deviation
- square root of the variance
Coefficient of Variation
- Measures the relative dispersion which expresses the standard
deviation as a percentage of the mean
Frequency Distribution:
- the proportion of subjects found in the different classes of the
population
Proportion, ratio and rate:
Proportion
- a / (a+b)
- e.g. proportion of patients cured
Ratio
- a/b
- e.g. sex ratio = no. of males : no. of females
Rate
- e.g. incidence rate = (no. of new cases over a period of
time/population at risk) x 100,000
How is the method of data analysis determined?
The method of data analysis that should be used for a study
depends on:
- the objectives and the research design
- the scale of measurement/type of data
METHOD OF DATA ANALYSIS
Descriptive Statistics
- Computations of indicators
Measures of central tendency: mean, median, mode
Measures of variation: range, standard deviation
frequency/percentage distribution; rates, ratios
Organizing, presenting & summarizing data in forms that
facilitate interpretation
- Inferential Statistics
o computations of indicators
o comparison of indicators
o test for significance of observed differences
o determination of degree of association
Criteria for selection of a statistical test:
• Level of Measurement of the Data
– Nominal
– Ordinal
– Interval/Ratio
• Specific objectives to be attained
• Types of samples
• Assumptions of the test
Purpose of doing the test:
• to determine if a sample could have come from a population with a
stipulated mean or proportion or from a population of some pre-
specified distribution
• comparison of two means or proportions
• comparison of more than two means or proportions
• determination of correlation or association between variables
Type of Samples
• Independent samples – the elements in one sample is not in any
way influenced by the characteristics of the elements selected in the
other sample.
• Paired or related samples – samples which are matched for certain
variables
Two kinds of statistics test
Parametric tests – involves numerical data, measured either on an interval
or ratio scale
Non-Parametric tests – involves data that are not truly numerical being
only on a nominal or ordinal scale
Statistical Significance vs. Practical Significance:
• Practical Significance refers to the benefits that may be derived
from the application of the knowledge or technology tested by the
study.
• Statistical Significance measures the probability of accepting a
wrong hypothesis or of rejecting a true hypothesis.