Chapter II: Research Methods
Scientific method is an empirical process in which a hypothesis is rigorously tested by careful experiment and observation.
Empirical investigation is a method of establishing new knowledge by making observations and drawing logical conclusions from obtained data.
Theory is a testable explanation for a set of facts.
Hypothesis, literally little theory, is a prediction of the result of a study which relates the variables can be falsified.
Operational definition is a part of an experiment or a study that specifies concepts and how they are to be quantified.
Independent variable is a part of an experiment that can be modified to produce results.
Dependent variable is the measured result of an experiment or a study that is produced by changing the independent variable.
Random presentation is a tool of a research psychologist in which he/she produces stimuli to the subject by chance alone.
Data is raw information collected from a study that will be analyzed later.
Replicate is to repeat the original experiment/study to ascertain whether its results are consistent and credible.
Experiment is a method of research in which the scientists directly control the conditions and modify the independent variable in order to produce an outcome.
Confounding/extraneous variables are things that influence the results of an experiment and are not accounted for in the procedure.
Controls are identical conditions a researcher imposes on his/her subjects during an experiment.
Random assignment is an experimental tool in which all possible subjects have equal probability of being put into a particular group.
Ex post facto is a quasi-experimental method in which subjects studied already have desired conditions.
Correlational study is a quasi-experimental method that uses statistics to determine the relationship between two variables. Note that correlation does not necessarily mean causation.
Survey is a quasi-experimental method that asks subjects specific questions.
Naturalistic observation is a quasi-experimental method in which subjects are observed in non-laboratory, i.e. natural, conditions.
Longitudinal study is a quasi-experimental method in which a chosen group of subjects is studied over an extended period of time.
Cross-sectional study is a quasi-experimental method in which a representative sample of a population is studied at one specific interval of time.
Cohort-sequential study is a quasi-experimental method in which a representative sample of a population is studied over a relatively short period of time.
Personal bias originates from the researcher’s beliefs and opinions and can affect his/her study’s results.
Expectancy bias originates from the researcher’s expectations and can affect his/her study’s results.
Double blind study is an experimental tool that prevents both the researchers and the subjects from knowing exactly what independent variable is being tested.
Frequency distribution is a measure of variability that shows how often each value in a set of data occurs.
Histogram is a bar graph that represents frequency distribution.
Descriptive statistics is a mathematical tool that characterizes the results of a study.
Inferential statistics is a mathematical tool that indicates the credibility of results from a study.
Mean is a measure of central tendency that is calculated by dividing the sum of all values within a set of data by the number of numbers.
Median is a measure of central tendency that is the number in the middle when all values within a data set are arranged in order of magnitude.
Mode is a measure of central tendency that is the number that occurs most often within a data set.
Range is a measure of variability that is the difference between the largest and smallest values of a data set.
Standard deviation is a measure of variability that is the average difference between the values of a data set and their (arithmetic) mean.
Normal distribution is a bell curve that represents the variability of values that occur in a data set.
Correlation is the relationship between two variables; when one changes, so does the other. Note that correlation does not necessarily mean causation.
Correlation coefficient is a number between -1 and +1 that describes the how strong the relationship between two variables are.
Random sample is a section of a population selected purely on the basis of chance.
Representative sample is a section of a population that accurately reflects that population.
Significant difference is a numerical value that indicates the probability that a result is due to chance.