PSYC1001 Lecture Notes - Lecture 16: Brain Damage, Internal Validity, External Validity
2 Jul 2018
School
Department
Course
Professor
16/04/2018 Research Design
Anecdotes are NOT science – they are merely stories – no time spent on alternative
explanations – you cannot distinguish theory from evidence - it is prejudged – only one
hypothesis, seek to confirm it, and then done
Need to test hypothesis
“I got better” anecdote – “I am living proof that X works” “I had illness Y, and was treated
with X, and then I got better. Therefore, it was X” “Where is your proof? Immune system is
still a thing
Case studies – trained to write down EVERYTHING, even if it doesn’t seem important – you
can begin to put the pieces together and get more details into the WHY – arise from an
earnest attempt to understand what is going on, such that all details, whether or not they
seem relevant, are recorded in a scientific manner
Correlational studies – study 2 things per person to see how they are related – when at least
2 variables are measured from each case/person with a view to calculating a relationship
between the variables – measurements are simply taken and nothing is
manipulated/controlled, therefore difficult to know the direction of causation or if there is
evidence of causation
When writing the correlation coefficients:
The sign indicates the direction of the relationship
The number indicates the strength of the relationship (higher the better)
Correlation DOES NOT MEAN causation
Control condition – the same as the experimental condition in every way, except the factor
you are investigating. If you test those conditions and still find a difference, only then can
you draw a strong conclusion about causation
Control condition – Rest, plenty of fluids, plenty of food
Experimental condition - Rest, plenty of fluids, plenty of food, gaming
Correlation findings e.g.
Rich people live near the ocean. This does not mean that people are more motivated to
earn money if they live near the ocean!
People we lock in detention centres are violent. No, if you imprison someone with their
family and children without committing a crime, it causes psychological problems
Use data to determine risk but it is not always accurate
To draw a causal inference, you need to manipulate independent variable to measure
dependent variable
Independent variable (manipulated by experimenter) – presumed cause; antecedent
(previous) cause; stimulus variable
Dependent variable (measured by experimenter) – behaviour of interest; response variable
In true experiment, randomly allocate participants to different conditions, pre-existing
opinions DOES NOT MATTER – all independent variables of interest are controlled and able
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Anecdotes are not science they are merely stories no time spent on alternative explanations you cannot distinguish theory from evidence - it is prejudged only one hypothesis, seek to confirm it, and then done. I got better anecdote i am living proof that x works i had illness y, and was treated with x, and then i got better. Correlational studies study 2 things per person to see how they are related when at least. The sign indicates the direction of the relationship. The number indicates the strength of the relationship (higher the better) Control condition the same as the experimental condition in every way, except the factor you are investigating. If you test those conditions and still find a difference, only then can you draw a strong conclusion about causation. Control condition rest, plenty of fluids, plenty of food. Experimental condition - rest, plenty of fluids, plenty of food, gaming. Correlation findings e. g. rich people live near the ocean.
