PSYC 523: Statistics & Research Methods Flashcards
ANOVA
- Stands for “analysis of variance”
- Statistical procedure used in inferential statistics
- ANOVAs test for significant differences among 2+ groups
- Also test for main/interaction effects of the independent variable on the dependent variable
Ex:
Clinical vs Statistical Significance
- Clinical significance that refers to when the results of a study are judged to be meaningful in relation to the diagnosis or treatments of disorders
- Statistical significance refers to the actual results of the statistical analyses that aren’t attributed to the operation of chance or random factors
Ex:
Construct Validity
- Degree to which a test or instrument is capable of measuring a concept, trait, or other theoretical entity
- The two main types of contract validity found in social science research are
- CONVERGENT validity (how well the measure correlates with other well-established measure of the same construct) and
- DISCRIMINANT validity (how much the measure does not correlate with unrelated measures)
- Tests should have BOTH convergent and discriminant validity in order to have high construct validity
Ex:
Content Validity
- Extent to which a test measures all facets of the subject matter or behavior that’s being studied
- Content validity cannot be measured empirically/statistically, but rather is assessed through logical analysis
Ex:
Correlation vs Causation
- Correlation describes the relationship (either positive or negative) between variable
- Causation refers to when changes in one variable bring about changes in the other variable(s) (i.e. cause and effect)
- A correlation is necessary to establish a causal relationship, but a correlation between variable DOES NOT assume that their is causation between the variables
- Only experimental studies can establish causal relationships
Ex:
Correlational Research
- A type of research design in which relationships between variables are simply observed without any control over the setting in which those relationships occur
- Correlational research does not contain any intentional manipulation of variables by the researcher
- The correlation coefficient can range from -1.0 to 1.0 and describes the strength and direction of the relationship (either positive or negative)
Ex:
Cross-sectional Design
- Research design in which groups, who differ by one key characteristic (i.e. age, developmental level, etc), are compared at a single point in time
- Typically used to determine the prevalence of a condition
Ex:
Dependent T-test
- Type of statistical procedure that compares the means of two related groups
- Used to determine whether or not there is significant difference between those means
Ex: relationship between “before” intervention scores vs “after interventions scores
Descriptive vs Inferential
- Descriptive statistics depict the main aspects of the sample data without inferring to a larger population (i.e. mean, median, mode, range, standard deviation within the sample)
- Inferential statistics are inferences about characteristics of a population to be drawn from a sample of data from that population, while controlling for error as much as possible
Ex:
Double-Blind Study
- Type of research design in which neither the participants nor the researchers knows which treatment/intervention participants are receiving until the study is complete
- Eliminates the possibility of researcher bias toward a participant or group
Ex:
Ecological Validity
- Refers to the degree to which results obtained from research are representative of conditions in the wider world
- Research designs with higher ecological validity are assumed to be more generalizable in life outside the confines of research/treatment
- This differs from external validity in that ecological validity describes how the results can be applied in real-life while external validity describes how results can be applied to people outside of the sample population but not necessary real life
Ex: doing a study in on the effects of alcohol consumption and then ask them to interact. To increase ecological validity they carry out the study in a bar
Effect Size
- Refers to the magnitude or meaningfulness of a relationship between two variables
- The larger the effect size, the stronger the phenomenon
- This is interpreted as indicating the practical significance of research findings
Ex:
Experimental Research
- Research design that utilizes randomized assignment of participants and a systematic manipulation of variables while all other variables controlled (or attempted to be controlled)
- The objective of experimental research is to draw a causal inference (i.e. any change in the dependent variable was due to the manipulation of the independent variable)
Ex:
Hypothesis
- An empirically testable proposition about some fact, behavior, or relationship that is usually based on theory
- The hypothesis states the expected outcome that will result from the research design’s conditions or assumptions
Ex:
Independent T-test
- Type of statistical procedure that compares the means of two independent groups
- Used to determine if the means are statistically different
Ex: “TERMS WITH EXAMPLES THE BEST” for example
Internal Consistency
- The degree of interrelationship of homogeneity among the items on a test, such that they are consistent with another and measuring for the same construct
- This is tested by dividing the items in half and scoring them separately, then correlating the town halves to check for the degree of consistency
Ex:
Internal Vaidity
- The degree to which a research study is free from flaws in its internal structure (i.e. lack of confounding variables)
- If so, this means the observed relationship between variables in the study is reflective of actual relationship between the variables
Ex: treatment for depression, didn’t allow for comorbid diagnoses so there wasn’t confounding variable. However this reduced ecological validity
Interrater Reliability
- The extent to which independent evaluators produce similar ratings from the same target person or object
- Typically used with more subjective measure to ensure the soundness of the operational definition(s) of the variable(s) in a study
Ex: “EXAMPLES BEST”
Measures of Central Tendency
- Refers to the values that attempt to describe a set of data by identifying the central position within the set of data
- The 3 main measures of central tendency ate mean, median and mode
Ex:
Measures of Variability
- how the spread of the distribution vary around the central tendency
- THREE primary measures: 1. Range 2. Variance 3. Standard Deviation
- Helps determine which statistical analysis you can run on a data set
Ex:
Nominal/Ordinal/Interval/Ratio Measurements
- Four types of scales of measure used in statistics
- Nominal measurements are used for categorical data
- Ordinal measurements are used for data that is ranked by individual or variable
- Internal measurements are used for measuring the degree of difference between values when an “absolute zero” is NOT possible
- Ratio measurements are used for measuring the degree of data when an “absolute zero” IS possible
Ex: When gathering data on a new client you ask them questions and describe their data in different terms. You ask their gender (nominal), age (interval), you give them a likert scale to rate how they feel today (ordinal) and how many panic attacks are you having a day (ratio)
Normal Curve
- Theoretical distribution in which values pile up in the center at the mean , median, and mode and falls off into “tails” at either end of the distribution
- Normal curve produces the familiar “bell-shape” when plotted
Probability
- Refers to the degree to which an event is likely to occur in a randomly sampled population
- As the probability (or p-values) increases, it becomes more likely that the result occurred due to chance as opposed to experimental manipulation
Ex:
Parametric vs nonparametric statistical analyses
- Parametric statistical analyses are based on the assumptions about the distribution of variables in the population that are being tested
- Nonparametric statistical analyses are used when the data being analyzed is does not meet the assumptions about distribution of the variables in the population
- Parametric statistics are typically preferred because they are more likely to detect statistical significance than nonparametric statistics
- Nonparametric statistics are used when the sample size is small and may not have symmetrical distribution.
Ex:
Quasi-experimental research
- A non-experimental research design
- The researcher cannot 1. randomly assign participants to conditions
- (2) Cannot generally control or manipulate the independent variable
- (3) And cannot limit the influence of extraneous variables
Ex: When doing research on heroin addicts, you don’t assign heroin addicts because that would be unethical. You use participants that are already addicted to heroin instead. The researchers are assessing if a new treatment is effective. Since you cannot randomly assign individuals to each group, this is considered quasi-experimental research
Random Sampling
- Process of selecting a sample of study participants from a larger potential group of eligible individuals
- Each person has the same fixed probability of being included in the sample
- Some chance procedure is used to determine who specifically is chosen for the sample
Ex:
Regression
- Statistical technique that’s used to determine how a variable of interest (i.e. dependent variable) is affected by one or more independent variables
- Typically used to predict values or estimate the effect of some exploratory variable on the dependent variable
Ex: levels of testoerone vs levels of aggressive behavior prediction
Sample vs Population
- Population is the entire group that a researcher want to draw conclusions about
- Sample is a smaller subset of that population that the researcher collects data from
- random sampling is the gold standard
- sample should be reflective of the population or the results may not be generalized to the population
Ex:
Scientific Methodology
- Used to conduct research in standardized way
- used to make observations
- gather data
- form theories
- test hypotheses
- and interpret results
Ex:
Standard Error of Estimate
- Measure of the accuracy of predictions made
- AKA standard error of the residuals: the difference between the expected/predicted dependent variable value and the actual dependent variable value
- Measures how the data points are spread around the regression line
- Standard deviation of residuals is the standard error of estimate
Ex: A researcher wants to find out if there is a relationship between social media usage and depression. They collect data and find that there is a positive relationship between the two variables. Next, they calculate the regression line. Next, they want to know how accurate the predictions made using the regression equation are, so they calculate the standard error of estimate by comparing the expected data (regression line) to the actual data collected.
Standard Error of Measurement
- CONFIDENCE INTERVAL
- Common tool in research and standardized testing
- Provides an estimate of how much an individual’s score would be expected to change on re-testing with same/equivalent form of test
- The scores over the infinite number of tests are averaged. This score is considered an estimate of the true ability/knowledge (T True score)
- The smaller the standard error of measurement SEM, the more precise the measurement
- Has in inverse relationship with the reliability coefficient
Ex:
Standard error of the Difference (2 sample t-test)
- The estimated standard deviation of the differences between the means of two independent samples
- This means it’s the estimate of error between the two groups
Ex: Anxiety between teenagers in the US and Germany. A researcher conducts a study on how caffeine affects test scores. They take the mean of scores from each group (with or without caffeine) and calculate the differences between the means. They then used the standard error of the difference to find the amount of error between the estimated and actual difference.
Standard error of the mean (single sample z-test)
- The standard error of the mean is the average of the deviations of sample means around the population mean
- Compares a random sample back to the population in the population mean AND standard deviation are known (as opposed to a single sample t-test, where the population standard deviation is NOT known)
- Comparison can be done before experimental manipulation to make sure the sample is representative, or it cane done after manipulation
Ex: A researcher creates a new test to measure intelligence, which they test on a random sample. Because they know the mean IQ and the SD of the population, they run a single sample z test by calculating the standard error of mean. This confirms that their sample is indeed representative of the population.
Standard Error of the Mean, estimated (single sample t-test)
- Standard error of the mean is the average of the deviations of sample means around the population mean
- The basis of a single sample t-test
- Compares a single random sample back to the population, where the population mean is know BUT THE POPULATION SD IS UNKNOWN, so it uses sample SD to estimate population SD (in contrast to a single sample z-test which is used when both population mean and population SD are known)
- Comparison can be done before experimental manipulation to make sure the sample is representative , or it can be done after manipulation
Ex: A researcher creates a new eating disorder measure, which they test on a random sample. Because they know the population mean but not the population standard deviation, they use the sample standard deviation to make an estimate. They run a single sample t-test by calculating the standard error of mean, estimated, which compares the random sample back to the population. They apply their SD from the sample to the population in hopes that their sample is representative of the population
Standardization Sample
- Population of individuals who have previously well-documented intelligence and/or achievement levels
- Used to standardize a new or revised instrument
- This is to assure that the new or revised measure is reliably measuring what it is intended to measure
Ex:
Statistical Significance
- Degree to which a result cannot be reasonably attributed to chance or random factors
- Result is measurably meaningful and the null hypothesis is rejected
Ex:
Type I and Type II Error
- Type I error occurs when the research results reject the null hypothesis when it is in fact true (i.e. false positive)
- Type II error occurs when the research fails to reject the null hypothesis when it is in fact not true (i.e. false negative)
Ex:
