Statistical Data Analyses

Only available on StudyMode
  • Download(s) : 66
  • Published : March 7, 2013
Open Document
Text Preview
Statistical Data Analyses

Graeme Ferdinand D. Armecin, MHSS

Outline of Presentation

  

Overview of Research Designs Functions of Statistics Sampling Principles of Analysis and Interpretation (with Computer Package) – Descriptive Statistics – Inferential Statistics Graeme Ferdinand D. Armecin, MHSS

Statistical Data Analyses

Purposes of Research Design




Exploratory/Descriptive Research design – Basic or fundamental in the research enterprise – What is going on? – Example: What is the percentage of the population enrolled in health insurance programs? Explanatory Research design – Why is it going on? – The purpose is to avoid invalid inferences – Example: What is the effect of enrolment in health insurance programs to the cost of hospitalization? Graeme Ferdinand D. Armecin, MHSS

Statistical Data Analyses

Essential Elements of Research
Theory (literature)

Statistics

Research empirical/what is observed Design

Statistical Data Analyses

Graeme Ferdinand D. Armecin, MHSS

Functions of Statistics
1.

2.

3.

Characterizing/exploring patterns of data to make it more meaningful Making generalizations about population parameter from sample statistic Finding associations or relationships between and among the gathered data Graeme Ferdinand D. Armecin, MHSS

Statistical Data Analyses

Sampling
Sampling refers to taking a portion of a population or universe as representative of that population or universe. An Important Fact About Sampling: Different results and hence different conclusions can be arrived when inference about a parameter is made on the basis of a sample. This is due to the fact that a sample is just one of the many possible samples that can be selected. Statistical Data Analyses Graeme Ferdinand D. Armecin, MHSS

Sampling Designs




Probability Sampling- sampling design in which every unit or member of the population is given an equal chance of being part of the sample Non-probability Sampling- sampling design where not every unit or member of the population is given an equal chance of being part of the sample Graeme Ferdinand D. Armecin, MHSS

Statistical Procedures in Research

Probability Sampling
Sampling Design Simple Random Sampling (SRS) Description Each element in the population has an equal chance of being selected into the sample. Advantage/s Easy to implement with the use of randomlygenerated numbers. Disadvantage/s Requires a listing of population elements. Takes more time to implement. Periodicity within the population may biased the sample and the results.

Systematic Sampling

Selects an element Less expensive of the population at than SRS the beginning with a random start and selects every kth element thereafter

Statistical Data Analyses

Graeme Ferdinand D. Armecin, MHSS

Sampling Design Stratified Sampling

Description Divides the population into subpopulations or strata and uses simple random on each strata. Population is divided into internally homogeneous subgroups and uses random sample to select the subgroup.

Advantage/s Researcher controls sample size in strata. Provides data to represent and analyze subgroups. Economically more efficient than simple random sampling. Easy to do without a population list.

Disadvantage/s Increased error will result if subgroups are selected at different rates. Expensive if strata have to be created. Lower statistical efficiency due to subgroups being homogeneous rather than heterogeneous.

Cluster Sampling

Multistage Sampling

Process includes collecting data from a sample using a combination of techniques defined above.

Reduces the cost.

Lower statistical efficiency.

Statistical Data Analyses

Graeme Ferdinand D. Armecin, MHSS

Sampling Design

Non-Probability Sampling
Description Advantage/s
Getting a sample from population elements which are readily available.

Disadvantage/s

Sampling by Convenience

The cheapest and Least...
tracking img