Module 1 Flashcards
Scientific Method
consists of learning assumptions, goals and procedures for creating and answering questions.
What are the 4 goals of science?
- Description - what happened - describe a behaviour and the conditions under which it occurred.
- Explanation - why it happened - causes of behaviour.
- Prediction - what will happen next?
- Control - how to make it happen.
What are the 4 approaches to understanding?
- Authority approach
- Analogy approach
- Rule Approach
- Empirical approach
Authority approach
seeking out knowledge from sources that are believed to be reliable and valid.
However, you can’t follow it blindly
Analogy approach
analogy between some new events and a more familiar understandable event.
The problem with it is that it is open to a number of interpretations
Rule approach
try to establish laws or rules that cover a variety of different observations. It can save time and effort but if followed blindly it can threaten the advancement of understanding.
Empirical approach
testing ideas against actual events - observing behaviour and drawing conclusions.
How can we achieve Causation?
In an experiment where one factor directly affects another factor.
It must demonstrate that changing the first thing produces a change in the second and that there is no other possible cause for the change in the second thing
What are the components of an experiment?
- population sample
- dependent variable
- operational definition
- reliability and validity
- bias
- floor and ceiling effects.
- data types and scales of measurement
Population Sample
Population - members of a specific group and defined by the purposes of the experiment.
Sample - relatively small subset of a population that is selected to represent the population.
Representative Sample - characteristics and behaviour of the sample reflect those of a population and it ensures generalisability.
How do you achieve a representative sample?
through random sampling:
Random sampling - select members have an equal chance of selection in an unbiased manner.
Descriptive Statistics
summarises the data collected from samples
Inferential Statistics
generalises the sample to the population
Dependent Variable
It is the measurement taken.
It is what you record - depends on what the participant does.
Operational Definition
Where in many cases there is no direct way to measure what you want so you have to think about 2 things:
- property of interest (PI)- what you are trying to measure
- dependent variable - a measurable value the must indirectly reflect the PI
so. ..the operational definition is the specification of how the property of interest will be measured.
Validity
Validity - a DV is valid if it measures what is is supposed too - threat to validity arises from any unintended component that is related in a score
Reliability
Reliability - a DV is reliable if, under the same conditions, it gives the same measures and contains a minimum of measurement error - unreliable data reflect error and provide a biased perspective.
Bias
a biased DV is consistently inaccurate in one direction - always high or always low
Floor and Ceiling Effects
Ceiling Effect - when a task is so easy that all the scores are very high.
Floor Effect - when a task is so difficult that all the scores are very low.
Data types and Scales of Measurement
the data type determines what sorts of analyses you can perform on your data and the conclusions you can draw.
either numerical (interval or ratio scales) or categorical - which goes into ordered (ordinal scale) or unordered (nominal scale)
Nominal Scale
categorises without ordering.
numbers that substitute for names.
e.g. 1 = female
Ordinal Scale
categorises and orders the categories.
Bigger means more.
Distance between points on the scale is not considered equal.
e.g. rugby team standings
Interval Scale
Categorises, orders and establishes and equal unit of measurement on the scale.
Knowing how much more.
Distance between points on scale is considered to be equal.
e.g. celcius temperature
Ratio Scale
Categorises orders and establishes an equal unit on the scale.
True zero point.
Allows ratio statements
Independent Variable
the experimental factor (s) that distinguishes your groups
Manipulated by the experimenter.
Has two or more levels.
What happens if you have a Quasi experiement
if you have a quasi experiment you MAY NOT conclude that there is a casual relationship between the IV and the pattern of results
Confounding Variable
If participants are not randomly assigned to condition, there may be differences between groups other than the DV.
this means you can draw conclusions that aren’t valid
Important issues regarding IVs
- Control groups
- Placebo effects
- Choosing levels properly
- single and double-blind studies
- demand characteristics
Control Group
Comparison group.
Differs from experimental groups by the absence of the experimental treatments
Placebo Effects
A special control group that believes it is receiving treatment when its not.
The performance in this group gives an estimate of the size of the placebo effect.
Single and Double-blind Studies
Single-blind = when a control group is used to measure the placebo effect - essential that the participant doesn't know what group they are in. Double-blind = neither experimenter or participants knows the condition that the participant is assigned to
Demand Characteristics
cues in a new situation that people interpret as “demands” for a particular behaviour.
either from participants, experimenter or environment = biases
Between Subject Design (IV)
Each participant is tested in only one level of the IV.
There may be confounds and to stop this use matching - which is a way to control extraneous subject variables - matched to treatment conditions.
Within-subject Design (IV)
each subject is test in every level of the IV - each participants serves as their own control.
This design reduces error variance which makes it easier to detect small systematic differences between treatment conditions.
What is the PROBLEM with within-subject design?
- Order Effects - order in which participants experience levels
Order Effects
Order in which participants experience levels can be a problem - can be fixed with counter-balancing.
Counter-balancing = each treatment condition is equally exposed to the practice effects and demand characteristics.
What is it important to control
subject variables, demand characteristics and experimental materials.
Multiple Independent Variables
More than 1 IV.
Allows determination of the effect of each IV on the DV and also how they interact.
Factorial Design
Where you have a number of IVs in your design and you collect data in all combinations of the levels of your IV your design is fully crossed.
What is the Main Effect of Factorial Design
the effects of one IV on the DV, ignoring the other IVs in the study.
there is a main effect for each IV
Interaction Effect
The effects of one IV on the DV taking into account the other IV(s) in the study.
In graphs: if the lines diverge or intersect then there is an interaction.
if the lines are parallel then there is no interaction
How to calculate midpoint
lower exact limit + 1/2 class interval width
Shapes of Frequency Distribution
uniform or rectangular distribution