Exam 1 Flashcards
Mode
The most frequent score in a distribution (the option with the highest frequency)
Range
The difference between the largest and smallest values
Median
The middle score when you place the scores in order from smallest to largest
Standard Deviation
The square root of the variance (used to measure the variation in a distribution)
Random Selection/Assignment of Subjects
- A type of assignment that maximizes chances that groups have similar characteristics at the start of a study
Purposive Selection/Assignment of Subjects
- Handpicking subjects who the researcher believes will lend insight into their study/research
Dependent Variable
The outcome of a study and variable that the researcher is interested in (must be measurable)
Skewed Data
A score that is so far from the mean on the normal bell curve that it ends up affecting the mean
Independent Variable
The variable that is believed to affect the outcome (DV). This is the aspect that the researcher manipulates (must have at least 2 variables
Repeated Factor
An IV (factor) that is repeated for each subject
Interval/Ratio Scale
The most quantitative measurement scale that can includes mathematical operations, order, and distance
Ex) weight, height, temperature
Test-retest Reliability
Reproducibility with tests results over time (The more consistent the test results are, the more test-retest reliability there is)
Internal Validity
The extent to which the results of the study demonstrate that a causal relationship exists between the independent and dependent variables
External Validity
Concerns to whom, in what settings, and at what times the results of the research can be generalized
Hypothesis
A proposed explanation made on the basis of limited evidence as a starting point for further investigation (prediction)
Directional Hypothesis
Experimenter predicts that a specific relationship exists between variables and predicts the direction (positive or negative) of that relationship
Non-Directional Hypothesis
Experimenter predicts that the there is a relationship that exists between variables, but does NOT specify the direction (+/-) of that relationship
(Null and Alternate Hypothesis are examples)
Complex Hypothesis
A hypothesis that includes MORE than 1 independent/dependent variable
Simple Hypothesis
A hypothesis that involves ONLY 1 independent/dependent variable
Face Validity
The weakest type of validity - determining the validity of something by only viewing its “face”
Ex) Determining that a tool appears to measure what it is supposed to so you determine that it is valid (no comparison to other tool)
Content Validity
The extent to which a measure is a complete representation of the concept of interest
(a final exam representing knowledge of the whole course content)
Construct Validity
Concerned with the meaning of variables (abstract theoretical principles) within a study
(the researcher might say that shoulder ROM determines shoulder function, but one might question that that those are 2 separate measures that do not affect each other)
Criterion-Related Validity
Extent to which a measure is systematically related to another measure (comparing a new instrument/measure to a gold standard)
Between Subject Design (Independent)
Different groups receive different levels of the IV
Within Subject Design (Dependent)
Same subjects receive all levels of the IV (pre-test and post-test)
Mixed Design
Designs with at least one within subject factor and one between subject factor
Normal Distribution
The bell curve. Symmetrical frequency distribution that is defined in terms of mean and standard deviation of data set
1 std deviation from mean = 68%
2 std deviation from mean = 95%
3 std deviation from mean = 99%
T-Test (t for 2)
Evaluating the difference between 2 INDEPENDENT samples
Paired t-Test (t for 2)
Evaluating difference between 2 DEPENDENT samples
1 Way ANOVA
BETWEEN Subjects - Evaluating difference of MORE THAN 2 INDEPENDENT samples
2 Way ANOVA
BETWEEN SUBJECTS - Evaluating difference of MORE THAN 2 INDEPENDENT samples with TWO IVs
1 Way Repeated Measures ANOVA
WITHIN SUBJECT - Evaluate difference of MORE THAN 2 DEPENDENT samples
2 Way Repeated Measures ANOVA
WITHIN SUBJECT - Evaluate difference of MORE THAN 2 DEPENDENT samples with TWO IVs
Type I Error
- Your statistical conclusion states that there IS a difference, but in reality there IS NO difference
- Probability of making type I error = ALPHA (reduce alpha to reduce probability of making this error)
Type II Error
- Your statistical conclusion states that there IS NO difference, but in reality there IS a difference
- Probability of making type II error = BETA (reduce beta to reduce probability of making this error)
Alpha Level
How much probability of drawing incorrect conclusion can be tolerated?
(Generally accepted Alpha Level is 0.05 or 5%) - due to a sampling error
Probability Sampling
Randomization at some point in the sampling process (simple, systematic, stratified, and cluster sampling are all examples)
Subject Attrition
A threat to INTERNAL validity. Participants lost from different study groups at different rates or for different reasons
Instrumentation
Threat to INTERNAL validity. Changes in measuring tools, improper calibration of instruments, unreliable human instrumentation (subjective)
Maturation
Threat to INTERNAL validity. Changes within the participant/subject that is due to the passage of time (could change the outcome/DV)
Correlation Coefficient
A statistical summary of the degree of relationship that exists between 2 or more measures (could be between different variables or repeated measures of same variable)
(Scale of .00 (very little) to 1.0 (strong) that determines strength of relationship)
Multiple Comparison Tests
The next step after null hypothesis is rejected in an ANOVA: comparing group means and pinpointing the differences (post-hoc AKA after the fact)
Use the Scheffe, Bonferroni, and Turkey HSD tests
Pearson Product Moment Correlation Coefficient (r)
The average of the cross products of the z scores for the X and Y variables (example: averaging the cross products of z scores of 6 month ROM and Gait Velocity)
r: Range of -1 to 1 where positive values has a positive correlation and negative value has negative correlation
Experimental vs. Non-Experimental Study
- Experimental: researcher manipulates an Independent Variable
- Non-Experimental: does not manipulate an IV (descriptions, analysis of relationships, and analysis of differences)
Correlation vs. Comparison
- Correlation - compares relative position of SAME individuals/groups on 2 DIFFERENT variables (margarine consumption in Maine positively correlates to divorce rates in Maine) –> does not equal causation
- Compare/Contrast - DIFFERENT individuals/groups are being compared on the SAME variable