Lecture 2 Pyb100 PDF

Dangers of misinformed practice - Bettelheim\'s theory of autism - Parents must be neglecting the children or mistreated for them to be autistic - Theory comes from his experience in a concentration camp and what he observed in behaviour from prisoner - he correlated...

Dangers of misinformed practice - Bettelheim\'s theory of autism - Parents must be neglecting the children or mistreated for them to be autistic - Theory comes from his experience in a concentration camp and what he observed in behaviour from prisoner - he correlated that with autistic behaviour - Frontal lobotomy - Synapses malfunctioning in frontal lobe causes severe depression, psychosis etc - Remove white matter in brain What are the problems with these approaches? - Bettelheim\'s theory of autism - Didn't use any factual evidence - opinion based - Didn\'t think about other variables that may impact the \"correlation\" - Posed himself as an expert and trusted his judgment -\> practice became harmful - Frontal lobotomy - Not based empirical data - Unethical - needed to warn people about the post-surgery effects - He wasn\'t an expert in psychology   - Limitations of intuition and unsystematic observation     What distinguishes science is the desire for evidence - Important to be critical   Research concepts: theory and hypotheses - Good research is grounded in theory - Theory: a systematic ways of organising and explaining observations - Different schools of thought promote different theories - Leads to new predictions that can be tested - Based on available evidence - Broad - Good theories can - lead to more predictions and testable and backed by research - Fits the known facts - Is falsifiable -\> can be proven wrong - Theory provides the framework for a hypothesis - Hypothesis: a prediction about the relationship between two or more variables - Specific Example: - Being ignored by others on social media has a negative impact on state self-esteem - How could we test this? - Survey: - Must measure all variables - self esteem & social media - Experiment - 2 groups - group chat scenario where 1 group gets response and the other doesn\'t and evaluate feelings   Doing research in psychology - Choosing a research design - Choosing a sample - Operationalising variables - Reliability and validity - Correlational and experimental research - Sources of bias - Ethics and integrity   Different research designs are commentary A diagram of a diagram Description automatically generated with medium confidence Naturalistic observation - Researcher carefully observes behaviour without intervening - Participant observation - Researcher is also a participant - Advantages - In-depth observation of behaviour in natural setting, not contrived - Can provide new insights - Disadvantages - Reactivity: difficult to remain unobtrusive   Case study - In depth investigation of an individual person/situation - Using interview, direct observation, records, psychological tests - Advantages - Can give starting point - Can provide rich, compelling data to support theory - Disadvantages - Representative of general population? - Subjectivity - investigators may see what they expect to see   Surveys - Use questionnaires or interviews to gather information about specific aspects of behaviour - Advantages - Can get info that isn\'t observable (data on difficult-to-observe behaviour) - Accessible - data from a large sample - Disadvantage - Self report data can be unreliable - Intentional deception, social desirability, response sets, reliance on memory   Correlational research - Looking for relationships among variables - E.g cyberostracism and self esteem - Useful for studying variables that the researcher can\'t manipulate - Personality, intelligence, age, sex - Can demonstrate that a relationship exists, but can\'t demonstrate [causality]   Experimental research: establishing causes - To establish causation, the researcher - Manipulates one variable - Independent variable (IV) - Randomly assign participants to condition - Measures its effect on another variable - Dependent variable (DV) - Holding all other variables constant   Population vs sample Population: - The *entire* group of people we are interested in studying Sample: - A *subset* of the population selected for study Random sampling: - Each member of the population is equally likely to be included in the sample - Least bias Representative sample: - Possesses the important characteristics of the population in the same proportions Convenience sample: - Recruit whoever you can; might not be representative - Most common in psych   Operationalising variables: - Turning abstract concepts into concrete variables that we can measure or manipulate - In a correlational study, all variables are measured - How would you measure these variables? - Cyberostracism - Self-esteem - For an experiment, we need to manipulate the casual variable (the IV) and measure the outcome (the DV) - How could we manipulate cyberostracism? - How could we measure self-esteem? Reliability and validity - Reliability - Does the measure produce consistent results? Test-retest reliability - Does the test give similar values if the same participant takes it two or more times Internal consistency - Different items that measure the same variable should produce similar answers - be consistent Inter-rater reliability - Two testers who rate the same person on the same variable, should give similar ratings to the participant q - Validity - Does it measure what it\'s supposed to measure? - Is our measure related to other measures with demonstrated validity? - Is our measure related to an outcome it should be related to? - E.g., does high performance on an aptitude test predict performance on this job?   Correlational research: Looking for relationships - Scatterplot - Plot one variable against another and look at the relationship - Correlational coefficient - Measures the direction and strength of the relationship between 2 variables - Can demonstrate that a relationship exists but can\'t demonstrate causality   Experimental research: establishing causes - To establish causation, the researcher - Manipulates one variable - Independent variable (IV) - Randomly assign participants to condition - Measures its effect on another variable - Dependent variable (DV) - Holding all other variables constant     Identify the variables e.g - Help is less likely in an emergency when there are several bystanders present. - Help DV, no. bystanders (IV)   - Heart rate and blood pressure will be higher after viewing a violent film sequence as opposed to a nonviolent film sequence. - Heart rate and blood pressure (DV), film sequence (IV)   - Heat stress leads to disorientation and confused thought. - Disorientation and confused though (DV), heat stress (IV)   Basic elements of an experiment: - State your hypothesis - Manipulate the independent variable - Measure the dependent variable - Try to eliminate sources of bias/alternative explanations for results - Draw conclusions   Holding other variables constant: - Experimental group - Exposed to the experimental condition of interest - Control group - Not exposed to the experimental condition, but otherwise treated identically to the experimental group - Random assignment - Participants are equally likely to be assigned to the experimental or control condition In a good condition, the only difference between the experimental and control groups is that the experimental group has been exposed to the independent variable   Sources of bias: - Participant: - Demand characteristics - Participants respond in the way they think the experimenter wants them to respond - Placebo effects - Participants\' condition improves because they believe the procedures will help them - Experimenter: - Expectancy effects - The experimenter\'s expectations and behaviour bring about change rather than the IV itself - To eliminate bias: - Single-blind study - Participant is unaware of the purpose of the study/which condition they are in - Double-blind study - Both experimenter and participant are unaware of purpose of the study/condition   - Confounding variables: - Another variable that could be producing the observed effect - Confounding occurs when something other than your IV varies systematically across conditions and is correlated with the DV - E.g. if the experimental group walked a dog and the control group sat quietly - IV = time with dog, DV = happiness, confound = activity level - Affects the internal validity of a study - Sampling bias: - Sample is not representative of the population as a whole - Affects the generalisability of the results