Flashcards in Quantitative Research Deck (43)
3 types of quantitative Research
3 types of RESEARCH METHODS
Mixed Methods research
In experiments, the researcher systematically manipulates the (which variable) ________ _______
Experiments are (H/L) _______ in internal validity but (H/L) ________on external validity.
• Quasi-experiments lack one or both of these characteristics of true experiments:
1- Random assignment to conditions
2- Systematic manipulation of the independent variable
TRUE/ FALSE In Experimental research : Covariates and extraneous variables can be controlled experimentally
Quasi-experiments are (H/L/M) ________ on internal validity and external validity.
FOUR KINDS OF QUASI-EXPERIMENTS: (not important)
- Person-by-treatment designs
• Natural experiments
• Nature and treatment designs
• Patched-up designs
Person by treatment experiment example
A researcher is interested in studying the effects of sex and mathematical aptitude on learning algebra via the traditional lecture
• The independent variables are sex and mathematical aptitude.
• The dependent variable is the overall grade in the algebra class.
NATURAL EXPERIMENTS or
"Acts of GOD"
___________use frequency-based techniques such as chi-square tests, counts, percentages, and proportions.
_________use nonparametric techniques based on ranks, such as the Spearman rank Correlation and the Mann-Whitney U test.
Ordinal Data—use nonparametric techniques based on ranks, such as the Spearman rank
correlation and the Mann-Whitney U test.
___________Use parametric statistical techniques like correlation, t tests, and analysis of variance (ANOVA)
Interval and Ratio Data
THE FOUR THINGS YOU CAN DO WITH STATS
You can use stats to solve problems of description.
• You can use stats to solve problems of differences.
• You can use stats to solve problems of relationships (not the Dr. Phil kind).
• You can use stats to solve problems of classification
___________is a variable that varies systematically with the independent variable and could exert an effect on the dependent variable.
There are three main kinds of confounds
- Person confounds
- Procedural confounds
- Operational confounds
___________ is a variable is held constant in a study, but which might represent a restricted
context under which some effect will be observed.
• Non-experimental research does not control for either of these, and thus is a _____ _____ ______
weak approach to research.
We deal with four kinds of variables in research:
- Independent variables
• Dependent variables
• Extraneous variables
Remember the null hypothesis says essentially that ___________________
there is no effect in the population.
The alternative hypothesis and the null hypothesis are mutually exclusive (only one can
be true) and exhaustive (you must either reject or fail to reject the null hypothesis).
Type I error__________
If you reject a true null hypothesis,
If you do not reject a false null hypothesis,
Both errors are bad, but statisticians usually consider a Type _____ error to be more serious
All things equal, the Big Four are related as follows:
The larger the effect in the population ____________
The more powerful the test
The smaller the effect in the population, _____________________to detect the effect.
The larger a sample you need
Lowering alpha makes your test _____________
more conservative and less powerful.
Increasing alpha makes your test _____________
less conservative and more powerful.