PA 12

• Measurement


  The heart of psychological testing as a scientific enterprise for the study of human behavior
• Measurement


  The use of certain devices or rules for assigning numbers to objects or events
• Psychological tests


  Carefully chosen samples of behavior to which a numerical or category system is applied according to some pre-established standards
• Psychological testing


  Extensive with psychometrics or psychological measurement, and is one of the primary tools for the science and practice of psychology
• The use of numbers in testing, requires us to delve into statistics
• Variable


  Anything that varies
• Constant


  Anything that does not vary
• Types of variables


  • Visible (e.g. sex, color of the eye)
  • Invisible (e.g. personality, intelligence)
  • Defined as to pertain to very small sets
  • Defined as to pertain to very large sets
• Discrete variables


  Variables with finite range of values
• Types of discrete variables

  • Dichotomous (true-false, heads or tails)
  • Polytomous (family size, vehicular traffic counts, baseball scores)
• Discrete variables can be tallied precisely and without error
• Continuous variables


  Have infinite ranges and really cannot be counted, they are measures that are more or less accurate approximations
• In psychological testing, the interest is almost always on continuous variables
• All measurements are prone to error
• Errors in measuring discrete variables arise only from inaccurate counting
• Errors in measuring continuous variables are inevitable because of the limitations of measurement tools
• Psychological tests as a measuring tool are subject to many limitations, hence margins of error must always be estimated and communicated along with test results
• Nominal scale


  Numbers are used instead of words, only identity or equality
• Ordinal scale


  Numbers are used to designate an orderly series, identity + rank order
• Interval scale


  Equal intervals between units but no true zero, identity + rank order + equality of units
• Ratio scale


  Zero means none of whatever is measured, all arithmetical operations are possible and meaningful, identity + rank order + equality of units + additivity
• Discrete data can be measured only with nominal data or with ordinal scales if the data fall in a sequence of some kind
• Continuous or metric data can be measured with interval scale or ratio scales if there is a true zero point
• Continuous data can be converted into classes or categories and handled with nominal or ordinal scales
• In psychological testing, most measurements are done using ordinal scales
• Descriptive statistics


  Use numbers and graphs to describe, condense, or represent data, summarize and describe the characteristics of a set (also called distribution) of scores
• Inferential statistics


  Use data to estimate population values based on sample values, or to test hypotheses, tell us how confident we can be in drawing conclusions or inferences about a population based on findings obtained from a sample
• Raw scores


  Unwieldy, do not convey much meaning
• Frequency distribution


  A table or graph that displays the frequency of various outcomes in a sample, summarizes the distribution of values in the sample
• Percent frequency = frequency / number of respondents x 100
• Cumulative percent shows the percentage of the cases that fell at or below each score, can be read as percentile rank scores
• Grouped frequency distributions


  Organize scores into intervals of a convenient size to accommodate the data, and the frequencies are listed for each interval instead of for each of the scores
• Pie charts or bar graphs


  For discrete or categorical data
• Histograms or frequency polygons


  For continuous or metric data
• Mean


  The arithmetic average of a group of scores, sum of all the scores divided by the number of scores
• Median


  The middle score that splits a distribution into half when the individual scores are arranged in order of lowest to highest
• Mode


  The most frequently occurring score in a distribution, useful when dealing with qualitative or categorical variables
• The choice of measure of central tendency should give a good indication of the typical score in the sample
• Unimodal distribution


  Has only one peak or mode, bell-shaped and symmetrical, mean, median and mode are the same
• Bimodal distribution


  Has two peaks or modes