Scientific Foundations Of Psychology Week Three Key Terms

Random Error

Individual variations or situational variations, such as fatigue

Systemic Error

Method variations, such as training of assessor or the way a question is posed

Reliability

The extent to which a score is consistent and dependable across time and between observers

• Reliability = True Score/ true Score + Error

Validity

The extent to which the score is consistent with theoretical expectations about how the construct should behave

Theoretical Construct

Thing we are trying to take measurement of

• Can't be directly observed

Split-half Reliability

Scores from one half of the measure correlate with the other half

Internal Validity

Extent to which you are able to draw the correct conclusions about casual relatonships between variables

External validity

Refers to the generalisability or applicability of your findings to the real world

Construct Validity

A question of whether you’re measuring what you want to be measuring

Face Validity

Refers to whether or not a measure “looks like” it’s doing what it’s supposed to, nothing more

Ecological Validity

The set up of the study should be closely approximate to the real-world scenario being investigated

Categories Of Citizen Power:

• Degrees of Citizen Power

• Degrees Of Tokenism

• Degrees of Participation

Degrees Of Citizen Power

Highest section highlights actual power and control. This section involved negotiation, sharing power and citizen action

Degree’s Of Tokenism

Counterfeit power, including informing people of their rights and responsibilities, consulting people on thoughts/opinions, and placating those who are unhappy. People are involved only to show they were involved

Degree’s Of Participation

No power. Illusionary form of participation that occurs when people are misled into believing they have power in a process that denies them any power. Attempt to make people think they are suffering due to their own circumstance instead of certain organisations

Three Levels Of Involvement That Can Be Conflicted

• Consultation

• Collaboration

• User-Controlled

Consultation

Members of public asked to give their views in assiatnce with informed decision making

Collaboration

Denotes ongoing partnership between academic researchers and contributions to make decisions about research

User-Controlled

Research that is controlled, directed and managed by service used and servic user organisations

Benefits To Involving Those With Lived-In Research

• Provide extra dimension to data analysis

• Alternative perspectives

• Identify relevant themes

• Check validity of experimentation

• Correct misinterpretation

Ways of involving people:

Use one person with lived experience to get involved within multiple stages of research]

- Use advisory group to look into multiple aspects of the experiment, multiple perspectives

- Might have people involved in research team that have experience

- Work alongside community partners who are actively involved in the project

Constructing A Hypothesis

Educated guess about what a study might find

- Built/contrusted based on what has veen found by past research looking at similar question

- Must be testable

- quite specific to it can set parameter on research design

- Hypothesis should be phrased as a statement, not a question

- Must be falsifiable

Repeatability

Whether the eperiment can be conducted again and get the same results

Agreement

The agreement between different researchers

Consistency

Can the same results. be obtained again using standadised testing

How To Improve Reliability

Improve instruments and tools overtime -> increasing number of items, clearly written items, objective scoring technique

- Standardize the test situation and clear instructions are provided

- Stating limitations and/or adjusting them: Who does this apply to and stating who it does apply to

Convergent Validity

How correlations between measures of the same trait: how much overlap between two concepts? - overlap,corrlation

Discriminant Validity

How much we can seperate/how much resolution there is between two different factors - dont want to measure self-esteem with agression - no correlation/no relation

Content Validity

evaluates how well an instrument (like a test) covers all relevant parts of the construct it aims to measure