PSYC 2250 Lecture Notes - Lecture 12: Structural Equation Modeling, Effect Size, Multiple Correlation
26 Jun 2018
School
Department
Course
Professor
PSYC2250 Introduction to Psychological Research
Chapter 12 Understanding Research Results:
Describing Variables and Relationships Among Them
Scales of measurement: A Review
Nominal scale:
- no numerical or quantitative properties; levels are simply different categories or groups
- Yes/no question; gender, blood type…
Ordinal scale:
- rank order the levels from lowest to highest, but the intervals between each order are not
equal
- 1.2, 2.2, 3.2…undefined
- sick vs. health; guilty vs. innocent; social class
Interval scale:
- intervals between the levels are equal in size; no absolute zero point
- Temperature, relative distance…
Ratio scale:
- have equal intervals plus a true zero point
- Number of items (telephone calls, people, births, deaths…)
- Income, size, price…
Sometimes is hard to distinguish ordinal or interval scale is being used
Analyzing the results of research investigations
3 basic ways to describe the results:
A) Comparing Group Percentage
- E.g. 70% of women …. And 78% of men….
B) Correlating Individual Scores
find more resources at oneclass.com
find more resources at oneclass.com
PSYC2250 Introduction to Psychological Research
- To describe whether and how the variables relate to each other in non-experimental designs
C) Comparing Group Means
- Compare the mean responses of participants in two or more groups
Frequency Distributions
Frequency distribution: indicates the number of participants who receive or select each possible
score on a variable
- E.g. how many students received a given score in an exam
Graphing Frequency Distributions
Pie Chart: divides a whole circle into slices that represent relative percentages
- Frequently used in applied research reports and in newspaper and magazines
Bar Graphs: uses a separate and distinct bar for each piece of information
- Commonly used for comparing group means
Frequency Polygons: use a line to represent frequencies
- Most useful when the data represent interval or ratio scales
Histograms: uses bars to display a frequency distribution for a quantitative variable
- Value along x-axis are continuous; Bars draw next to each other
- Contrasts with bar graph, use when x-axis are nominal categories
Descriptive Statistic
Descriptive statistic: allow researchers to make precise statements about data
Central Tendency
Central tendency : tell what the sample as a whole, or on the average is like
3 measures: mean, median, and mode.
Mean: a set of score is obtained by adding all the scores and dividing by the number od scores
- Appropriate when scores are measure on interval or ratio scale
Median: the core that divides the group in half
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Nominal scale: no numerical or quantitative properties; levels are simply different categories or groups. Ordinal scale: rank order the levels from lowest to highest, but the intervals between each order are not equal. 1. 2, 2. 2, 3. 2 undefined sick vs. health; guilty vs. innocent; social class. Interval scale: intervals between the levels are equal in size; no absolute zero point. Ratio scale: have equal intervals plus a true zero point. Number of items (telephone calls, people, births, deaths ) Sometimes is hard to distinguish ordinal or interval scale is being used. 3 basic ways to describe the results: comparing group percentage. And 78% of men : correlating individual scores. To describe whether and how the variables relate to each other in non-experimental designs: comparing group means. Compare the mean responses of participants in two or more groups. Frequency distribution: indicates the number of participants who receive or select each possible score on a variable.
