PHTY211 Lecture Notes - Lecture 2: Scatter Plot, Glasgow Coma Scale, Central Tendency
Research Design - Basic Concepts 2
• Describe attributes & variables
o Attribute
• An inherent personal characteristic
• May have different values called variables
• Subjects are categorised or compared according to this attribute. E.g., gender, age
o Variable
• A characteristic of interest that can be measured or counted
• “oeties alled fatos
• Called a variable because its value can vary over time or within different groups
• Cause and effect relationship
▪ Is X related to Y?
• Types
▪ Independent variable (IV)
▪ Dependent variable (DV)
▪ Extraneous/confounding variables
• Differentiate between independent, dependent, extraneous variables
o Independent variable (IV)
• Usually represented by X
• Is manipulated by researcher so effect can be measured
▪ Treatment, e.g. effect of stabilisation exercise program
• Don't rely on other variables for their changes
• Has the presumed effect on the dependent variable?
• Must have at least 2 values (levels
o Dependent variable (DV)
• Usually represented by Y
find more resources at oneclass.com
find more resources at oneclass.com
• A response, an outcome, or effect
• The thing we want to effect a change on
• Values depend on changes in the independent variable
• So DV is the thing that we measure
▪ e.g. muscle strength, pain levels
• Must specify how to measure outcome
o Extraneous/confounding variables
• Factors other than those directly examined in the experiment that might influence
the relationship between the IV and the DV
▪ e.g. training, past injury history, environment
• Need to be controlled
• Failure to control extraneous variables affects internal validity of research
• Describe what is meant by inclusion and exclusion criteria
o Filter the participants entering a study/clinical trial
o Are there to ensure patient safety during study, ensure only appropriate subjects are
included, minimize withdrawal, ensure goals of study are reached.
o Maintain ethical approach to research
o Help to control confounding variables
• Describe populations and samples
o Study a part (sample) in order to gain information about the whole (population)
o Selecting a group of people, events, behaviours or other elements with which to conduct
a study
• Sample is a subgroup of the population of interest
o How accurate this is, depends on the extent that the sample population represents the
target population (external validity)
find more resources at oneclass.com
find more resources at oneclass.com
o Participant selection
o Must possess all characteristics the investigator has identified as necessary
• Inclusion criteria
o Must not possess any characteristics the investigator has defined
• Exclusion criteria
o Must be available for selection into the sample
• Differentiate between probability and nonprobability sampling
o Probability sampling
• Every person has the same probability of being selected from the target
population
• Samples are created through process of random selection
• Less sampling error results projectable
• Types of probability sampling
▪ Simple random sampling
• Unbiased
• Participants are allocated a number
• Random drawing of numbers from a hat or from a computer-
generated list of numbers
▪ Systematic sampling
• Beginning from a random position on a selected named list
• E.g. to select a sample of 100 students from a list of 1000, every 10th
person on the list is selected starting at a random point.
▪ Stratified random sampling
• Involves identifying relevant population characteristics and separating
members of a population into homogenous, non-overlapping subsets
based on these characteristics
• E.g. investigating attitudes of physiotherapy students to working
with the elderly but want to take into account level of
experience.
• Draw a sample of 40 physiotherapy students with equal
numbers from different stratified groups
• 10 first years + 10 second years + 10 third years + 10
fourth years
▪ Cluster sampling
• The researcher selects groups or clusters, and then from each cluster,
the researcher selects the individual subjects by either SRS or
systematic sampling
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Research design - basic concepts 2: describe attributes & variables, attribute, an inherent personal characteristic, may have different values called variables. Subjects are categorised or compared according to this attribute. E. g. , gender, age: variable, a characteristic of interest that can be measured or counted, o(cid:373)eti(cid:373)es (cid:272)alled (cid:862)fa(cid:272)to(cid:396)s(cid:863, called a variable because its value can vary over time or within different groups, cause and effect relationship. Independent variable (iv: dependent variable (dv, extraneous/confounding variables, differentiate between independent, dependent, extraneous variables. Independent variable (iv: usually represented by x. So dv is the thing that we measure: e. g. muscle strength, pain levels, must specify how to measure outcome, extraneous/confounding variables. Factors other than those directly examined in the experiment that might influence the relationship between the iv and the dv: e. g. training, past injury history, environment, need to be controlled. Samples are created through process of random selection. Less sampling error results projectable: types of probability sampling.
