Unit 1

• Biostatistics - theory and application of statistical science to analyze public health problems. 

  • Bio: life
  • Statistics: collection, organization, analysis, and interpretation of numerical data
• Uses of Biostatistics

  • compare two sets of data
  • compute the average, median, standard deviation
  • obtain a conclusion
  • find association between the two variables
  • find the correlation between the variables
  • give the results in a tabular or diagrammatic form
• Areas of Biostatistics

  • Health Statistics - in public health or community health
  • Medical Statistics - in Medicine; includes defects, injury, disease, efficacy of drug, serum and line treatment
  • Vital Statistics - in population; study of vital events like births, marriages and deaths
• Role of Statistics in Research 

  Aids the researcher in:

  1. Designing a research project 
  2. Processing, organizing  and summarizing research data 
  3. Quantifying variability
  4. Interpreting results and drawing valid conclusions
• Why do we need statistics?

  • Variation 
  • Tendency of a measurable characteristic
  • change with respect to person, place & time
  • E.g. weight, age, height, etc.
  • Necessary to analyze variability in order to
  • Describe certain characteristics or make valid conclusions. 
• Population

  • all subjects/ samples of interest
  • Result: Parameter
• Samples

  • selected subjects/ samples of interest
  • Result: Statistics (estimate)
• Parameter

  • the subject of interest
  • describes the whole population
• Statistics

  • an estimate of a parameter
  • number describing a sample
• Variables:

  • Characteristics of interest to be measured.
  • The value that varies from one individual to another or within the same individual to different periods of time.
  • Most of the time, variables are dependent on constants
• Constant

  • Fixed characteristics.
  • Numbers that do not change.
  • Most of the time they are imposed by the researcher themselves.
• POPULATION

  • Entire group of individuals or items of interest in the study.

  TARGET POPULATION

  • Group from which representative information is desired, and to which inferences will be made.

  SAMPLING POPULATION

  • Population from which a sample will actually be taken.
• SAMPLING UNIT

  • Units chosen in selecting the sample, and may be made up of non-overlapping collection of elements.
• ELEMENTARY UNIT OR ELEMENT

  • Person or object on which measurement is actually taken or an observation is made.
  • Example: to determine the prevalence of asthma in 1st year college student:
• SAMPLING FRAME

  • Collection of sampling units.
  • Tool that allows for drawing a sample (ex: listing, spot maps, aerials photographs).
  • The availability of the sampling frame determines whether or not there is a gap between the target and sampling population.
• SAMPLING ERROR

  • The difference between the value of the parameter (i.e. the true value) and estimates of these values based on different samples.
  • Deviation of sample values compared to true sample values
  • Parameter – statistics
• Data

  • Raw material of statistical data
  • Figures or figures result from the process of counting
  • Figures from taking a measurement

  Examples:

  • Number of patients in the hospital
  • Number of laboratory tests to be analyzed
• Variable

  • a characteristic that takes on different values in different persons, places, or things
  • For example: 
  • heart rate, 
  • the heights of adult males, 
  • the weights of preschool children, 
  • the ages of patients seen in a dental clinic
• Quantitative - can be measured in the usual sense

  Qualitative - many characteristics are not capable of being measured. Some can be ordered (ordinal) and cannot be ordered (nominal)
• Quantitative Variables

  1. Discrete
  2. characterized by gaps or interruptions in the values that it can assume. 
  3. whole numbers
  4. Continuous
  5. an assume any value within a specified relevant interval of values assumed by the variable. 
  6. usually decimal in reports
• Types of Qualitative

  1. Nominal - As the name implies it consist of “naming” or classified into various mutually exclusive categories
  2. Ordinal - Whenever qualitative observation Can be ranked or ordered according to some criterion.
• Statistical Methods

  • The collection, presentation, analysis, and interpretation of numerical data

  1. Collection of Data - the first step in data collection
  2. Presentation of Data - the mass data collected should be presented in a suitable form for further analysis
  3. Interpretation of Data - the final step is drawing a conclusion from the data collected
• Sampling:

  • act of studying or examining only a segment of population (or sample) to represent the whole
  • Whatever findings we obtain from the sample, we generalize to the total population
• Importance of Sampling

  • Cheaper and Faster
  • Better quality of information collected
  • More comprehensive data collected
  • Only possible method for destructive procedures
  • More “ethical” especially in intervention studies
• • Population entire group of individuals or items of interest in the study
  • Target Population Group from which representative information is desired, and to  which inferences will be made
  • Sampling Population population from which a sample will actually be taken
• Ideally,

  TARGET POPULATION = SAMPLING POPULATION
• TARGET POPULATION ≠ SAMPLING POPULATION

  • Unavailability of information for sampling purposes
  • Inaccessibility of the target population

  Example:

  • Research Objective To determine the prevalence of learning disabilities among children aged 7 to 12 years
  • Target Population - all children aged 7 to 12 years old
  • Sampling Population all school children aged 7 to 12 years old
• Definition of Terms

  • Sampling Unit - Units chosen in selecting the sample, may be made up of non-overlapping collection of elements
  • Elementary Unit or element Person or object on which measurement is actually taken, or an observation is made
• • Sampling Frame
  • Collection of sampling units
  • Tool that allows for drawing a sample (e.g. listing, spot maps, aerials photographs)
  • The availability of the sampling frame determines whether or not there is a gap between the target and sampling population
  • Sampling Error
  • The difference between the value of the parameter (i.e. the true value) and estimates of this values based on different samples
  • Parameter- Statistics
• Criteria of a Good Sampling Design

  • Sample should be representative of the population
  • Sample size should be adequate
  • Sampling procedure should be practical and feasible
  • Sampling design should be economical and efficient
• Factors to Consider in Selecting/Developing the Sample Design

  • Nature of the variables
  • Population being study
  • Purpose for which the research undertaken
  • Availability of information relevant to sampling procedure itself
• Judgment or Purposive Sampling

  • A sample is selected based on an expert’s subjective judgment or on some pre-specified criteria

  Example:

  • Selection of study areas based on:
  • Proximity
  • Level of cooperation of community members
  • Familiarity of investigators with community leaders
• Accidental or Haphazard Sampling

  • Sample is selected based on whatever items come first or whoever is available

  Example:

  • “Person-on-the-street” sampling
  • “Ambush” interviews
• Quota Sampling

  • Selection of items or individuals to include in the sample takes place until a pre-specified number (quota) is reached

  Example:

  • Patient satisfaction survey is conducted until the “required sample size” is completed
• Snowball sampling

  • Frequently used when studying hidden populations (e.g. IVDU , PLWA)
  • Devised because of the difficulty producing the sampling frame and in identifying members of these populations

  Procedure:

  • The first person identified to be a member of the target population is interviewed / included in the study
  • He / She will be asked to identify the next person for inclusion in the study
  • And so on…
• Simple Random Sampling

  • Most basic type of probability sampling design
  • Every element in the population has an equal chance of being included in the sample
  • Chronologically-numbered listing of the population is required as the sampling frame
  • Often used in studies involving a relatively small population, with readily available sampling frames
• Systematic Sampling

  • Variation of simple random sampling
  • Involves calculating for the sampling interval (k)
  • Every kth unit is included in the sample
  • Appropriate sampling design when:
  • A sampling frame is not available
  • Sampling units are too numerous to list for purposes of simple random sampling
• Stratified Random Sampling

  • Stratified Random Sampling
  • Involves dividing the population first into overlapping groups (or Strata)
  • Simple random sampling is then carried out within each stata
  • Appropriate sampling design to:
  • Ensure that sub-groups considered important in the research are adequately represented
  • Increase the precision of estimates of parameters being considered (more so when values of variables considered are heterogeneous but homogeneous within the stratum)
• Cluster Sampling

  • Sampling design where the sampling units are clusters of elements
  • Sampling unit ≠ Elementary unit
  • Appropriate sampling design when:
  • Sampling frame for elements is not readily available
  • Cost consideration
• Multi-stage Sampling

  • Appropriate for sample surveys that have a wide coverage
  • Advantages
  • Complete sampling frame is necessary only for the first stage of selection, while the sampling frame for succeeding stages of collection are confined to those selected as sample
  • Disadvantages
  • Increase in the variance estimate
  • Complexity of data analysis