Why do we need Statistics?
Psychology is an empirical science.
Researchers gather information
Statistics help researchers make sense of the information and data that they gather
Organize & summarize
Communicate to others
Decide whether conclusions are justified based on results obtained
Psychological Research
Two elements:
1.) Research Methods: make observations & formulate questions. The architect of the study.
2.) Statistics: the tools that you use to answer those questions. The contractor of the study.
Definition of Statistics
A set of methods and rules for organizing, summarizing, and interpreting information.-----data
“ . . . order out of chaos”
“a set of standardized techniques that are recognized and understood throughout the scientific community”
Descriptive statistics:
Summarize, organize, simplify data (focus for chapters 1-5) ex. Average of something
Inferential statistics:
Methods that use sample data to make general statements about a population
Used to draw conclusions about a population
Populations vs. Samples
Population: set of all individuals of interest in a particular study
Sample: set of individuals selected from the population of interest, which are intended to represent the population
Variables = characteristic or condition that changes or has different values for different individuals; also environmental variables
Data (plural) = measurements or observations
Datum (aka raw score)
Data set
Scales of Measurement
Nominal Scale:
(A set of categories)
Ex: Academic majors, gender, name of cookies
Ordinal Scale:
(Categorized ranked observations)
Ex: freshman, sophomore, junior, senior, birth order, book edition ranking of favorite cookies
Interval Scale:
Ordered categories are intervals of exactly the same size
Arbitrary zero point, therefore can’t determine exact differences between scores
Ex: shoe size, temperature of cookies (no such thing as no temp.)
With interval data, we can add and subtract, but cannot multiply or divide. Confused? Ok, consider this: 10 degrees + 10 degrees = 20 degrees. No problem there. 20 degrees is not twice as hot as 10 degrees, however, because there is no such thing as “no temperature” when it comes to the Celsius scale.
Ratio Scale:
Interval scale with an absolute zero point
Ratio scales allow you to compute the actual difference between two scores
Ex: height, weight, exam score, How many cookies are left?
Parameters | Statistics |
data/value derived from a population
| data/value that describes a sample
typically, any population parameter has a corresponding sample statistic |
Discrete variables | Continuous variables |
|
|
Sampling Error:
Estimated error between the data obtained from a sample and the population intended to study
Example: GPA of this class will not be exactly the same as GPA of all college students in US…
Variables in Research
Independent
That you manipulate or categorize
Dependent
That you measure; it depends on the independent variable
Confounding
Systematically varies with the independent variable and so you try to control or randomize away
Constructs
Definition: hypothetical concepts that are used to help define behavior
Have to define a construct to be measured
Operational definition:
defines a construct in terms of an observable and measurable response
Examples: Anxiety, depression, IQ, self-esteem
Selecting and Assessing Variables
Operational definition
Exactly what you are studying
Reliability
Consistency of the measure
Validity
Extent the test measures what it is supposed to measure
Hypothesis Testing
The process of drawing conclusions about whether a relation between variables are supported or not supported by the evidence
Correlational Method | Experimental Method |
NOTE: correlation does NOT equal causation More on this method later . . . | -Determines cause and effect Experiments are…
Examining a cause and effect relationship
|
One Goal, Two Strategies
Between-groups designs
Different people complete the tasks, and comparisons are made between groups
Within-groups designs
The same participants do things more than once, and comparisons are made over time
An example of the experimental method
Quasi-Experimental Method
Comparing groups which were NOT created by manipulating an independent variable
Groups are determined by a participant variable
Real limits
Definition:
Boundaries of intervals for scores represented on a continuous number line
Limit separating 2 scores is located exactly halfway between them
Each scores has 2 real limits
Upper real limit
Lower real limit
Scores:
X = scores for a variable
N = no. of scores in the population
n = no. of scores in the sample
Sigma:
Σ = summation X = 20 13 7 9 8
ΣX = 57, n = 5