Quantitative Research Methods Flashcards
Evidence based practice
Best research evidence
Clinical expertise
Patient characteristics and values
Descriptive statistics
Condense a large amount of information into smaller pieces (summary) of information
Inferential statistics
Statistical information about a population from a sample of that population with a calculated degree of confidence
Test between different groups
T-test
Analysis of variance
Test relationships between variables
Correlation
Regression
Compare 2 groups
T-test
Compare 2 or more groups
ANOVA
Correlation
Explore the relationships between pairs of variables
Bivariate regression
Predict scores on one variable from scores on another variable
Multiple regression
Predict scores on a dependent variable from scores on a number of independent variables
Descriptive statistics
Frequencies
Percentages averages
Assumptions in statistics
An assumption is a condition that ensures that what you are attempting to do works
Nature of data
Continuous or categorical
Categorical data
Categories of data are best presented and interpreted with bar graphs
Continuous data
Data that can be measured on scale which can interpret median, mode and mean from
Population
Collection of units to which we want to generalise a set of findings or a statistical model
Sample
A smaller collection of units from a population used to determine truths about that population
The only equation you will ever need
Outcome=(model) + Error
Mean
The value from which the scores deviate least.
Type 1 error
Occurs when we believe that there is a genuine effect in our population when in fact there isn’t
Type II error
Occurs when we believe that’s there is no effect In The population when in fact there is
One-way analysis of variance is used when
You have only one independent variable (eg. gender)
Two-way analysis of variance is used when
You have two independent variables (gender, age group)
Independent variable
The proposed cause
A predictor variable
A manipulated variable
Dependent variable
The proposed effect
An outcome variable
Measured not manipulated
NOIR
Nominal
Ordinal
Interval
Ratio
Binary variable
There are only two categories
Eg. dead or alive
Nominal variable
There are more than two categories
Ordinal variable
The same as a nominal variable but the categories have a logical order
Eg. Fail, pass, merit, destinction
Interval variable
Equal intervals on the variable represent equal differences in the property being measured. Cannot have a 0
Ratio variable
Similar to interval variable but can have a 0 baseline
Categorical variable
Binary
Nominal
Ordinal
Continuous
Interval
Ratio
Measurement error
The discrepancy between the actual value we’re trying to measure and the number we use to represent that value
Validity
Whether an instrument measures what it set out to measure
Content validity
Evidence that the content of a test corresponds to the content of the construct it was designed to cover
Ecological validity
Evidence that the results of a study, experiment or test can be applied, and allow inferences, to real-world conditions.
Reliability
The ability of the measure to produce the same results under the same conditions
Test-retest reliability
The ability of a measure to produce consistent results when the same entities are tested at two different points in time
Correlational research
Observing what naturally goes on in the world without directly interfering with it
Cross-sectional research
This term implies that data come from people at different age points with different people representing each age point
Experimental research
One or more variable is systematically manipulated to see their effect (alone or in combination) on an outcome variable.
Statements can be made about cause and effect.
Between-group/between-subject/independent data collection
Different entities in experimental conditions
Repeated measures (within-subject) data collection
The same entities take part in all experimental conditions.
Economical
Practice effects
Fatigue
Null hypothesis Ho
There is no effect
The alternate hypothesis H1
Aka the experimental hypothesis
Statistical statement format
Statistic
Degrees of freedom
Value
Sognificance
Effect size
Dependent samples t-test
Repeated measures design.
Whether the same group of individuals differ on a particular measure.
Before-After design
Analysis of variance (ANOVA
Evaluate mean differences between two or more treatments or populations
ANOVA key terms
Independent Variable= factor
Treatment (condition) of a factor = level
ANOVA study with more than one factor
Factorial design
2 factors = two-way factorial design
4 factors = four-way factorial design
(Eg. Exercise: yes/no, personality: type A/B, type of meds: Panadol/ibuprofen, education level achieved: university/secondary
One-way independent measures ANOVA
Independent measures design.
A seperate sample is taken for each level. (Eg. Age groups:suggest ‘mutual exclusivity’)
Repeated measures design: one sample of individuals are in both levels of treatment condition. (One boys heart rate taken once before and once after running a race)
ANOVA decides if
Differences between the sample means represent real differences between the treatments. That is the treatments really do have different means and the sample data accurately reflects those differences.
There really is no difference between the treatments. The observed differences between samples are due to chance.
Statistical hypothesis for ANOVA Ho
states there are no differences between the populations represented by the treatments
Statistical hypothesis for ANOVA H1
The population mean for at least one treatment means is different from others
F statistic
Simultaneously compares all sample means in a factor to determine whether two or more sample means differ significantly
F statistic Formula
F= between groups variance
————————————
Within groups variance
Between groups variance
Two possible explanations:
Treatment (experimental) effect.
Chance.
- individual differences
- experimental error
Within treatment variance
Cannot occur because of a treatment effect! But can occur because of:
Chance
- individual differences
- experimental error
F= treatment effect+ differences due to chance
——————————————————
Differences due to chance
When the treatment has no effect, then the differences between treatments (numerator) are entirely due to chance.
If the differences are due to chance, the numerator and the denominator should be approximately equal and the F-ratio should have a value around 1.
ANOVA F statistic/formula: when the treatment does have an effect, causing differences between the samples
The between treatment differences (numerator) should be larger than the chance (denominator).
A large F-ratio indicates that the differences between treatments are greater than chance.
The treatment does have a significant effect.
ANCOVA
Tests whether the IV still effects the outcome variable after the influence of the covariants has been removed