The viewing and recording of a predetermined set of behaviors
Observational research
Does NOT incorporate external validity
Does NOT incorporate ecological validity
External validity
The extent to which study findings can be generalized outside the data collection setting to other persons, in other places, at other times
Ecological validity
The degree to which the research situation re-creates the psychological experiences that participants would have in real life
Coding system
A set of rules to help guide how the researcher classifies and records the behaviors under observation
Duration recording
Recording the length of time a target behavior occurs
Frequency-count recording
Recording each time a target behavior occurs
Observation schedule
A hard-copy document, or electronic form, on which observers note the particulars of the behavior or phenomenon they are observing
Inter-observer reliability
The level of agreement between two observers coding the same phenomenon, also known as inter-rater reliability
Intra-observer reliability
The extent to which a SINGLE observer consistently codes a phenomenon
Reliability goal: acceptable agreement
Pilot testing
A trial run used to test and refine the design, methods, and instruments of a study prior to carrying out the actual research
Continuous recording
A procedural method for recording observations that involves recording all the behavior of a target individual during a specified observation period
Interval recording
A procedural method that involves dividing the observation time into smaller intervals and recording the occurrence or non-occurrence of the target behavior during each interval
Contrived observation
Artificially introducing a variable of interest and unobtrusively observing the outcome
Descriptive statistics
Statistics that describe or summarize quantitative information
Categorical variable
A variable with a finite number of different values
Continuous variable
A variable with an infinite number of different values between two given points
Frequency distribution
A graphical depiction of the distribution of scores in a data set
Normal distribution
Represents the ideal distribution of scores in a population
Symmetrical
Basis of statistical testing
Kurtosis
Refers to the distribution of scores relative to the middle
Skewed distribution
Only one pronounced tail
Measure of the degree of asymmetry
Mode
The most frequently occurring score in a distribution
Bimodal
Having twomodes, more difficult to analyze
Median
The score at the 50th percentile (50% of the scores at or below the median)
Calculating the median
1. For small N, discrete data with odd number of scores: median is the middle score
2. For small N, discrete data with even number of scores: median is the halfway point between the 2 middle scores
Mean
The most commonly used measure of central tendency
Refers to the score located at the mathematical centre of a distribution
Measures of variability
Describes the extent to which scores in a distribution differs from each other
Range
The distance between the 2 most extreme scores in a distribution
Variance and standard deviation
Measures to describe differences between scores
Based on the normal distribution (with respect to the mean)
Score deviation
The "distance" between an individual score and a sample mean
Calculating unbiased estimate of population variability
Use N-1 degrees of freedom
Standard deviation and area
68% of the scores in a distribution are within of + 1SX of the mean
Coefficient of variation
Used for interval or ratio data
Variation as a proportion of the mean
Unitless - independent of the units of measurement
Normal distribution and z-score
Assess relative position of each unit in the sample
Eliminates unit of measurement
Percentages and proportions
Frequency - counting the number of times a score occurs
Calculating percentage
Percentage = (f/n) x 100%
Calculating proportion
Proportion = f/n
Ratios
Compares the relative sizes of categories, i.e., part to part
Rates
Number of actual occurrences of some phenomenon divided by the number of possible occurrences per some unit of time