Term
|
Definition
Proof of causation requires that the following conditions are met:
1. Time ordered effect
2. Proper spatial relationship
3. be able to rule out other possible causes
Philosophically, it is not possible to empirically show causation.
In the practical world, we rely on the weight and preponderance of evidence to make decisions. |
|
|
Term
|
Definition
| Assignment of symbols to objects, events, and concepts (something that we can give a number and value to that concept) |
|
|
Term
| Properties of a good measure ('score') |
|
Definition
1. Observable (empirically grounded
2. Representative - of that which is measured
3. Valid (e.g., equivalent to the true score) |
|
|
Term
|
Definition
| Classification of objects(groups), mutually exclusive groups, groups are labeled with symbols. |
|
|
Term
| Nominal Level: Mathematical Properties |
|
Definition
Counts (frequencies)
Percentages
Mode |
|
|
Term
|
Definition
| All nominal level properties plus: groups (categories) can be ordered |
|
|
Term
| Ordinal Level: Mathematical Properties |
|
Definition
|
|
Term
|
Definition
| All ordinal level properties plus: there is an equal distance between all measurement scale points (distance between 50 and 100 is the same as the distance between 400 and 450). There is a set distance between numbers. |
|
|
Term
| Interval Level: Mathematical Properties |
|
Definition
|
|
Term
|
Definition
| All interval level properties plus: the ratio between scores has meaning, there is an absolute zero (most of the time things will be Ratio if we are going to best classify the measure) |
|
|
Term
|
Definition
Ratio measures are interval measures with an absolute zero.
Interval Ratio
Temperature Height
pH Weight
Age
Blood Pressure |
|
|
Term
| Ratio Level: Mathematical Properties |
|
Definition
|
|
Term
| Present the Strengths and Limitations of Measurement Levels |
|
Definition
Information Hierarchy
i.Nominal = lowest level of measurement
Yields the least information
ii.Interval (ratio) = highest level of measurement
Yields the most information
iii.Don’t waste information!
Use the highest level of measurement possible
iv.Important in literature evaluation and reporting results |
|
|
Term
| Changing the Measurement Level |
|
Definition
| May change the results (e.g., mask effects by "adjusting" the measurement scale) |
|
|
Term
|
Definition
| Those that are varied by and under the control of the experimenter. |
|
|
Term
|
Definition
| Those that respond to experimental manipulation. |
|
|
Term
| Continuous vs. Discrete Variables |
|
Definition
Continuous variables
a. Partial units (e.g., decimal parts) make sense
i. Weight = 60.7 kg
ii. Average cost = $52.35
Discrete variables
a. Partial units are not found naturally
i. Average family size is 3.2 |
|
|
Term
|
Definition
Study subjects compare objects and put them in order (e.g., rank products by taste)
-Advantage
Direct comparisons are a familiar task
-Disadvantage
Difficult to order more than three or four items
"Objects" can be anything
-Ex. physical products, patient attributes, etc. |
|
|
Term
|
Definition
| There are different types of rating scales, but all are used to judge how much of a trait an object (real or abstract) posesses |
|
|
