HPS111 Lecture Notes - Lecture 1: Biomed Central, List Of Fables Characters, Google Scholar
Seminar- notes
WEEK 2:
Interpreting results:
Results- section where the plot points come to a climax. Tell us the output from our method
Discussion- and the part where the characters narratives are tied up nicely (or atleast enough for
now). Interpret the results, tell us how they answer our question (or not), tell us what this means
and doesn’t mean, tells us whats next.
Results-
Presents study findings without interpretation
Describe and communicates the analyses:
- Written descriptions/explanations
- Statistically analyses (if they are true associations/analyses)
- Tables and figures (major results)
Intro to stats
Statistics are a set of techniques to:
- Describe, or summarize a sample- descriptive statistics
- Make judgments about a population- inferential statistics
Descriptive stats:
- Central tendency (means, medians, modes)
- Standard deviation
Inferential stats:
- Stats for drawing conclusions about “the population” ie given what we can see, what can
we know about the world
- An astronomer, a physicist and a mathematician are on a train in Scotland
- There are lots of different types of inferences/questions, each with different tests to
stop us drawing hasty inferences
- Two main types:
oDifferences- are groups different from one another?
Do men and women differ on jealousy? T-test
Does high school type predict 1 years grade, ANOVA or f-test
oRelationship/associations- are variables related? What’s the relationship?
find more resources at oneclass.com
find more resources at oneclass.com
Is time spent studying related to exam grade? Correlation r and
regression
For each % increase in average score at uni, how much does lifetime
earning increase? Assuming it does, regression
Statistical significance
P<0.5 means the data are surprising if there is truly no difference/relationship
Probability less than 5% then it is very likely there is a relationship/difference
Correlation is not causation
Significance means the effect is unlikely to be illusory, it does not mean things are causally linked
Reading the results
What questions are being asked?
Frequency of reports of tails, vs oxytocin and competitive environment (difference and relationship)
What tests were used?
Tail filp recorded
What was found?
Oxytocin would increase dishonesty, competitive environment would increase dishonesty, people
are quite dishonest (what we want). More dishonest in the competitive environment. Placebo there
was less likely to be dishonest. The oxytocin group competitive environment was much higher.
There was no partial liars there was a 33% increase to be a full liar. Beliefs about lying rate increase
likely hood to lie. They tend to conform more in a competitive environment.
Shame and your behaviour of lying. More shame in non-competitive environment. Don’t feel as
much shame in competitive environment. Less shame in non-competitive environment when they
belief that others are lying too
Understanding of discussion
- Summaries study aim and confirms/rejects hypotheses
- Compares study findings to previous research
- Explains relevance and implications
- Describes limitations (threats to internal (trust)/external (population at large)validity)
- Provides suggestions or future research
- Summaries overall contributions to literature
What did we find?
The effect of oxytocin has a detrimental effects on honesty in an competitive environment.
find more resources at oneclass.com
find more resources at oneclass.com
Inclination to conform with their peers, dishonest behaviour, which indicates that the detrimental
effects of oxytocin on dishonesty in competition were driven by conformity
Ocytocin showed no effect on the inclination to lie when competition was not present
How did this fit within the broader literature?
Given oxytocin more likely to conform like other studies found
Compared to other studies saying they found similar
What limitations were there?
Weren’t included – sample is all male (only apply to male), young age
What implications are there?
Whats next?
WEEK 3:
Answers:
Questions
1. What predictions (hypotheses) were made?
That the educational handout would increase the effectiveness of placebos
End of the introduction, placebo will reduce the pain
1. Describe the measures of interest
Placebo, education, cold pressure task, shock intensity (ranked pain threshold- intense ethic)
2. Did the authors compare differences or relationships?
Relationship between education and perceived enhancement. Both comparing differences in
placebos groups. And relationships across variables. Education v non, placebo v non.
Relationship between. How much they expected the placebo to work, how familiar they
were with anorgesics.
3. Did the authors report significant/non-significant findings?
Significantly less pain. They did find significant finds for the electric shock, 605 80% 100%, no
significant at 60% but yes for the others, p<0.5. they found also, important, at 80% the
education was enhancing the placebo.
4. Were the hypotheses support or rejected?
Partially supported. For the most part. It wasn’t happening for the cold pressure educed
pain.
5. Were the findings consistent with previous studies?
Yes. That is consistent, but partially because there was no placebo effect under all conditions
6. What does the study contribute to the area?
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Results- section where the plot points come to a climax. Discussion- and the part where the characters narratives are tied up nicely (or atleast enough for now). Interpret the results, tell us how they answer our question (or not), tell us what this means and doesn"t mean, tells us whats next. Make judgments about a population- inferential statistics. Stats for drawing conclusions about the population ie given what we can see, what can we know about the world. An astronomer, a physicist and a mathematician are on a train in scotland. There are lots of different types of inferences/questions, each with different tests to stop us drawing hasty inferences. P<0. 5 means the data are surprising if there is truly no difference/relationship. Probability less than 5% then it is very likely there is a relationship/difference. Significance means the effect is unlikely to be illusory, it does not mean things are causally linked.
