sumit kataria

Research in Disability Rehabilitation: Basic Concepts:

Spring 2007 Research in Disability Rehabilitation: Basic Concepts Presented by Sumit Kataria

Slide 2:

Spring 2007 Variables Variables are things that vary and change

Types of Variable:

Spring 2007 Types of Variable Independent Variable Dependent Variable Confounding Variable

Independent Variable:

Spring 2007 Independent Variable Something that is change by the Researcher What is tested What is manipulated

Dependent Variable:

Spring 2007 Dependent Variable Something that might be affected by the change in the Independent Variable What is observed What is measured The data collected during the investigation

Confounding Variable:

Spring 2007 Confounding Variable Any variables that can potentially play a role in the outcome of a study but which is not part of the study

Type of Confounding Variable:

Spring 2007 Type of Confounding Variable Two type of Confounding Variable Extraneous Variable Intervening Va riable

Extraneous Variable:

Spring 2007 Extraneous Variable Any variable other than the independent variable that could cause a change in the dependent variable

Intervening Variable:

Spring 2007 Intervening Variable Other factor that arises during the course of research

Slide 10:

Spring 2007 For Example:

Slide 11:

Spring 2007 “Impact of Programmed Instruction on Pre-Arithmetic Skills amongst Primary students with Mild Mental Retardation”

Independent Variable:

Spring 2007 Independent Variable Programmed Instruction

Dependent Variable:

Spring 2007 Dependent Variable Pre-arithmetic skills

Confounding Variable:

Spring 2007 Confounding Variable Extraneous Variables: Socioeconomics Status, level of motivation of students and subjects . Intervening Variables: Type of Programmed Instruction, tools of data collection.

Hypothesis:

Spring 2007 Hypothesis Tentative statement on solutions to the problem Tentative statement relationship between two and more variables An educated guess of conditions of a phenomenon A reasoned speculation about how two and more variables are related to each other

Criteria of Research Hypothesis:

Spring 2007 Criteria of Research Hypothesis Should explain expected relationship that exist between two or more variables. Researcher should have strong reasons based on concrete evidences or theories to formulate the hypothesis which is to be tested. Should be as short as possible but clear. Should be testable.

Types of Research Hypothesis:

Spring 2007 Types of Research Hypothesis Null Hypothesis (H O ) Alternative Hypothesis (H A )

Example of Research Hypothesis:

Spring 2007 Example of Research Hypothesis There will be no significant difference on the over all performance of the students before and after intervention. There will be no significant difference on each domain performance of the students before and after intervention.

Slide 19:

Spring 2007 Sampling

Population:

Spring 2007 All individuals in a group that has similar characteristics (one or more) to be studied by the researcher . Population

Sample :

Spring 2007 Part of a chosen population to be observed and analyzed By observing the randomized samples’ characteristics, several inferences on the population may be made Sample

Slide 22:

Spring 2007 Parameter Values obtained form a population Statistics Values obtained from a sample

Types of Sampling Method :

Spring 2007 Probability Sampling Methods Non-Probability Sampling Methods Types of Sampling Method

Probability Sampling Methods:

Spring 2007 Simple random sampling Systematic sampling Stratified sampling Cluster sampling Multi-stage sampling Probability Sampling Methods

Simple random sampling:

Spring 2007 Elements are chosen at random so that each element has an equal chance of selection Simple random sampling

Systematic sampling:

Spring 2007 The first element is chosen at random. Subsequent elements are chosen using a fixed interval Systematic sampling

Stratified sampling:

Spring 2007 The population to be sampled is divided into homogenous groups based on characteristics. Stratified sampling

Cluster sampling:

Spring 2007 A simple random sample of clusters is chosen from a sampling frame. Then, simple random sample of individuals within each cluster is selected. Cluster sampling

Multi-stage sampling:

Spring 2007 This is like cluster sampling, but with several stages of sampling and sub-sampling. This method is usually used in large-scale population surveys Multi-stage sampling

Non-probability Sampling Methods:

Spring 2007 Convenience sampling Quota sampling Snowball sampling Non-probability Sampling Methods

Convenience sampling:

Spring 2007 A sample is drawn on the basis of opportunity. For example, the sample could include youth attending a school activity, service providers attending a conference or parents attending a school event. Convenience sampling

Quota sampling:

Spring 2007 A sampling frame is defined in advance of data collection and a sample is chosen from this list, but not at random. Quota sampling

Snowball sampling:

Spring 2007 Data is collected from a small group of people with special characteristics, who are then asked to identify other people like them. Data is collected from these referrals, who are also asked to identify other people like them. This process continues until a target sample size has been reached, or until additional data collection yields no new information. Snowball sampling

Advantages Disadvantages of Probability Sampling Methods:

Spring 2007 Advantages Disadvantages of Probability Sampling Methods Advantages Less prone to bias Allows estimation of magnitude of sampling error, from which you can determine the statistical significance of changes/differences in indicators Disadvantages Requires that you have a list of all sample elements More time-consuming More costly No advantage when small numbers of elements are to be chosen

Advantages Disadvantages of Non-Probability Sampling Methods:

Spring 2007 Advantages Disadvantages of Non-Probability Sampling Methods Advantages More flexible Less costly Less time-consuming Judgmentally representative samples may be preferred when small numbers of elements are to be chosen Disadvantages Greater risk of bias May not be possible to generalize to program target population Subjectivity can make it difficult to measure changes in indicators over time

Questions:

Questions

Thank You:

Thank You