Slides Week 2 Flashcards
Research attempts to . .
- Increase Understanding
- How and Why do we behave the way we do.
How does research start
- noting an interesting question
- stating the question in a way that it can be answered Undergoing the scientific Method
There are different types of research design (2)
- Non Experimental
- True Experimental
- Quasi Experimental
- Experimental
- Descriptive
- Historical
- Correlational
- Qualitative
How are the research types different
- nature of the question asked
- method used to answer questions
- degree of precision of the method
Non Experimental Research
- describes relationships between variables
- cannot test cause-and-effect relationships
- descriptive
- historical
- correlational
- Qualitative
Descriptive Research
- describes characteristics of an existing phenomena
- provides a broad picture
- serves as basis for other types of research
Historical Research
- Describes past events in the context of other past or current events
- Primary and secondary sources of data
Correlational Research
- Asks what several events have in common
- Asks whether knowing one event can allow prediction of another event
- Does not imply causation
Qualitative Research
- Asks what several events have in common
- Asks whether knowing one event can allow prediction of another event
- Does not imply causation
Types of Research Design (Table)

What is the difference between a variable and a value?
- A variable is a factor that can be measured
- A value is a subset of a variable
Eg: height is a variable, 186cm is a value
Independent Variable
- A group or condition in a study
- Is what we are measuring
- Divided into levels
- Directly or indirectly manipulated by researcher
- Direct Manipulation: drug treatment
- Indirect Manipulation: school grades
What makes a good IV?
- Not confounded
- IV Levels do not vary systematically with other variables
- DV is sensitive to changes in the IV
- Called Dependant because its “Scores’ depend on experimenter manipulation
Dependant Variable
- The thing being assessed or measured
- Measures outcome or performance
- Example: Amount of time looking at screens (IV); level of health and fitness (DV).
- Needs to be operationalised
Operationalised
- Clearly defined IV & DV
- Specific description of how you will define and measure a variable
- define as it is used in your study.
Control Variables
- Variable whose influence you want to control
Ie: sex difference in thrill seeking behaviour you may control for income
Extraneous Variables
- Confounding occurs when an extraneous variable either:
- Varies systematically across levels of IV
- Is correlated with the DV
Define True Experiment
- Participants randomly assigned to groups
- Treatment variable is controlled by researcher
- control of potential causes of behaviour
Quasi-experiment
- Participants are assigned to groups
- useful when researcher cannot manipulate variables
What is a variable?
- An entity that can be measured and can take on different measured values
eg: height, weight, intelligence, hair colour, time, performance, income, level of depression
Moderator Variables
- Also called mediator variables (although there is a difference between them)
Mediator Variable
- Is thought to describe the psychological process that occurs to create the relationship
Moderator Variables
- Change the strength of an effect or relationship between two variables
Dependent Variable Definition and AKA
A variable that is measure to see whether the treatment or manipulation of the independent variable had an effect
AKA: Outcome variable, Results Variable, Criterion Variable
Independent Variable Definition and AKA
- A variable that is manipulated to examine its impact on a dependent variable
AKA: Treatment, Factor, Predictor Variable
Control Variable Definition and AKA
- A variable that is related tothe dependent variable, the influence of which needs to be removed
AKA: Restricting Variable
Extraneous Variable Definition and AKA
- A variable that is related to the DV or IV that is not part of the experiment
AKA: Threatening Variable
Moderator Variable Definition and AKA
- A variable that is related to the DV or IV and has an impact on the DV
AKA: Interacting Variable
Between Subjects Design
- Also known as an indepedent sample
- each subject is exposed to one level of each IV
Within Subjects design
- Also known as a repeated measures design
- Each subject is exposed to all levels of each independant variable
Define Hypothesis
- “if . . . then” statements
- objective extension of the original question
- in a testable form
- hypotheses posit a relationship between different factors
- data collected that will confirm or refute the hypothesis
- hypotheses are testable not provable.
Hypotheses are . . .
- Brief, declarative statement
- predict the outcome of a study
- Posed as a priori
- Are well educated guesses
Define a Priori
A complete well written Hypothesis should . . .
- be stated in declarative form
- posit a relationship between variables
- reflect a theory or body of literature upon which:
- they are based
- be brief and to the point
- be testable
Why do we need hypotheses?
- Karl Popper
- Falsification
- Hypotheses can be falsified/rejected
Falsification
The process by which something can be demonstrated to be false
Why is falsification important?
- Important to Poppers philosophy and how scientific knowledge progresses
- Hypotheses are refined and subsequent theories are developed
- Hypothesise develop strength and credibility
*
The Null Hypothesis
- A statement of no difference/relationship/effect
- nothing is going on
- the starting point for evaluating research
- Evaluating research assumes null to be true
- then attempts to collect evidence to knock that down.
H0: μ1 μ2
The Alternative Hypothesis
- A statement that something is going on
- Demonstrates relationship, effect or difference
- Can be non directional or directional
- depends on level of confidence - but is very important
- Non Directional: H1: μ1 ≠ μ2
- Directional: H1: μ1 > μ2
Populations
- A collection of units (people, cats, plants)
- What we wish to generalise in our research findings
- Populations large or narrow
- We aim to infer about general populations
Samples
- A smaller colleciton of observations from a population (people, cats, plants)
- used to infer characteristics about the population
- Bigger samples are more likely to be accurate
- results may vary across samples
- on average will be similar
Testing the whole population
- It is expensive and time consuming to test an entire population
- Instead we use samples: Mini Populations
- Randomly selected samples from a population can be said to reflect the entire population.
- Based on samples, we can generalise to the population
Field’s Model Analogy
The Real World
→
1) Good Fit
2) Moderate Fit
3) Poor Fit

Descriptive Statistics
Aim to capture the essential features of the results in an easily comprehensible form.
- Violent Video Group: Average aggressive acts = 9.49
- Neutral Video Group: Average violent acts = 6.22
But: Surely when we run our experiment we need more than this to test our hypothesis?
Statistical Significance
- How do we know if the difference is large, small or a chance result?
- there are Three Important questions
- Is there a statistically significant difference between the two groups
- what do these results tell us? Do they support, or fail to support hypothesis?
- are these sample results an accurate reflection of the population
Esitmation or Inference
- When we test a hypothesis statistically we want to reach a conclusion.
- What is going on in our population of interest
- We use sample data to give us information to form a conclusion
What is a p value . . . ?
- Statistical significance testing allows us to test for differences between groups.
- Also allows us to test relationships we observe
- a process that tells us if we can reject the null hypothesis or if we can retain it.
- Significance level = risk associated with not being 100% certain that null hypothesis is incorrect
- Calculated as a p value
P Value
- A hypothesis test that is used to determine the significance of the results from a study.
- It is the probability that the results from an experiment or study are due to chance and not the experimental conditions.
- also known as calculated probability
- P value only tells you about probability; not meaningfulness.
Null Hypothesis Significance Testing
- Most used method for testing research questions with statistics
- Fisher and the Lady Tasting Tea
- claims that we should calculate the probability of an event
- then evaluate this within the research context
*
What is a population?
- a collection of units we want to generalise in research findings
Why do we use descriptive Statistics?
To capture the essential features of the results in an easy to comprehend form
What is the most common method for testing research questions with statistics
Null-hypothesis significance testing.