1

Science and Statistics in PsychologyLecture 1

Theory and EvidenceTheory and Evidence

Dr Caleb Owens

Brennan MacCallum 453

caleb.owens@sydney.edu.au

Consultation: Wednesdays 9-10am

2

Important notes about this lecture series• Pseudo-sciences are used throughout this lecture series as examples to

emphasize the care and precision which science requires– There is no malign intent involved in choosing particular examples of

pseudoscience. If you find yourself challenged by a particular example, you will find dozens more given in lectures which can serve your understanding equally well. The main purpose of these lectures is to give you the toolkit to criticize. If you find that you can be highly critical of some things in your life but not others – that’s fine, and a curious psychological anomaly we know well (even if we don’t understand it). The specific aim here is to encourage you to have a strong sceptical attitude toward all your subject areas in psychology. Psychology is a new science, and is particularly hard for us to research objectively because it is about us, so a high level of scientific rigour and understanding of science is necessary.

– Any form of knowledge needs to be used responsibly. It is not recommended you use the knowledge you gain from this series (or any other lecture series) to mock or deride any outsider. As students sitting voluntarily in a science lecture, you are assumed to be open to the implications of science. However in the outside world, not everyone holds the same world view, and an insensitive approach can cause great offence and hurt.

– If you wish to contribute to or start a debate on any of the issues raised in this lecture series, the webCT forums for PSYC1001 are open and waiting. Use what you learn in this lecture series to make your arguments persuasive, for example, refer to appropriate evidence (peer reviewed research rather than anecdotes) and avoid logical fallacies. Everyone gets to contribute to science, but not every hypothesis can be right.

• The items posted on webCT are lecture overheads only. On many slides however you will notice one or two ‘notes’ at the bottom which clarify or reinforce particular points. Keep an eye out for them.

3

By the end of this lecture series:• You should have a clear idea what distinguishes science from

pseudoscience, and be able to use this understanding to help advance psychology

• You should feel confident in your ability to criticize and evaluate almost all kinds of psychological research for its validity and importance

• You should have a good understanding of the main issues which confront psychologists attempting to both design and interpret research

• You should understand that statistics and a grasp of it is central to all psychology

• NB: This is NOT a mathematics course– If you have taken a “statistics” course in mathematics and feel you can

skip these lectures, you are much mistaken– Equally, if you are afraid of mathematics you will be surprised at how

little maths there is– Understanding statistics in psychological research is an exercise in

empowerment, logic, and scepticism – there is very little messing around with numbers, there is no calculator required for the final exam.

4

Lecture Plan

1. Theory & Evidence : Science and pseudoscience; the importance of a rationale

2. The power of a name : Measurement and constructs3. Predictions : Hypotheses in science; the null hypothesis;

the importance of a disprovable hypothesis4. A thousand zeros : Types of research design; internal

and external validity5. The coin toss : Understanding variability in sampling

and measurement; probability and the appeal to ignorance

6. Too much of a good thing : Statistical power; Practical significance

5

What is science ?• An attitude – a quest for understanding

• People of science want to make sure they are not mistaken • Continuous doubt, uncertainty, reflection, reanalysis and

scepticism• Humility about current understanding, always hoping for

something better

• A technique for discovering the nature of the world

• Science has taken a long time to develop and requires training

• Structures of self-correction which (eventually) result in valid knowledge

• Scientific structures like peer reviewed journals, research funding mechanisms play an important role

6

Current understanding of the world

Criticism forbidden

No progress whatsoever

7

Current understanding of the world

Analysis and criticism of current theory

Hypotheses proposed about other explanations

Hypothesis #1 Hypothesis #2 Hypothesis #3

Not a better explanation

Not a better explanation

A better* explanation

*But not necessarily perfect.

New theory proposed

8

Science is power• No other system of understanding the

world comes close

• Have you heard of:– The Eastern Science moon project with chi

powered rockets?– An alternative chakra powered mobile phone

network?• At most you may have heard of ‘alternative

understandings’ in fields with unclear results, and small, hard to replicate placebo mediated effects (e.g. “alternative” medicine)

9

What science is good at

• Explanations– Which answer questions and solve problems

• And do not lead to more questions about the same problems

• Predictions– Of how substances will react, structures will

perform, drug effects, technology, addictive behaviour, consumer behaviour

10

What science is bad at• Purposes

– So we understand how to split the atom… what do we do?

• Make nuclear weapons to destroy each other?• Build nuclear power plants?• Not build nuclear power plants?

• Meaning and ethics– What was the point of all those people who

died being born in the first place?– Is it right/good to conduct experiments on

animals?– Science might tell us whether/which animals can feel

pain though

11

Is psychology a science?

• Psychology is a challenge: a large series of problems and issues concerning how we understand behaviour.

• At most we could say that we want to study the challenges psychology presents us with in a scientific manner

12

Theory and evidence

• Science needs both:– Evidence without theory could be considered

‘blind empiricism’. We can record a vast number of phenomena which we cannot fully explain but progress is difficult

– Theory without considering evidence is often a complete waste of time

• How theories and the evidence which supports them interacts is the key to the success of science

13

Theory – mechanisms, constructs, causal relationships

Evidence – findings from controlled experiments, anecdotes, testimonials, historical records, chemical evidence etc.

Scientific Knowledge – Established theories and facts, Previously replicated evidence

Is the theory supported by evidence?

Does the theory

conflict with

established

knowledge?

Does the evid

ence

conflic

t with

previo

us

findings?

14

What is a theory?

• A model of how something works– E.g. the Copernican model of the revolution of

the planets around the sun

• A metaphor for the functioning of a system– E.g. the body as electrical (1820

Frankenstein)

• An interrelation of concepts which is supported by observations and which leads to new predictions

15

Some key theories in science

• Plate tectonic theory (from late 1960’s)– Preceded by:

• Geosyncline theory (mid 1800’s)• Continental drift (1912)

– Key evidence:• Continents ‘fit’ together• Age of the earth (based on cooling)• Magnetic anomalies• Seismic imaging

– Used to predict and explain• Sea floor spreading, plates colliding forming mountain ranges• Earthquake and volcano zones

16

• Germ theory of disease– Preceded by:

• Spontaneous generation of disease• Evil agents inside the body (curses, negative thoughts)• Bodily imbalance (of forces, humours, energies)• Vitalistic concepts (energy, intention, purpose)• Miasmatic theory of disease (caused by pollution)

– Key evidence:• Microscopes developed which could detect agents

– Discovery of viruses and bacteria• Koch's (1890) postulates:

– The microorganism must be found in abundance in all organisms suffering from the disease, but should not be found in healthy animals.

– The microorganism must be isolated from a diseased organism and grown in pure culture.

– The cultured microorganism should cause disease when introduced into a healthy organism.

– The microorganism must be reisolated from the inoculated, diseased experimental host and identified as being identical to the original specific causative agent.

– Used to predict and explain• Effectiveness of antibiotics, hygiene, sterilization• Led to development of vaccines, treatments, prevention

17

• Modern evolutionary synthesis– Preceded by:

• Lamarckism – Acquired traits passed on (led to Lysenkoism in Russia)

• Orthogenesis (single direction change, e.g. bigger, more complex)• Saltation (distinct from punctuated equilibrium)• Evolution by natural selection

– Variation in offspring leads to differential reproduction• Genetic drift

– Random changes in allele frequency may remove alleles when selective pressures absent (e.g. cave animals lose pigment)

– Key evidence:• Mendelian genetics

– Selective breeding of dominant and recessive genes• Embryology• Discovery of the structure of DNA• Palaeontology and geology• Modern genetics and genetic mapping (including population

genetics)– Used to predict and explain

• Effectiveness of antibiotics, drugs• Origins of species (futures of species)• Development and function of animal parts including the human brain• Ecosystems and ecology

18

• (in Psychology) Localization of function– Preceded by:

• Thinking with the heart• Phrenology / Reflexology / Iridology

– Key evidence:• Brain damage• Lesion studies• Brain scanning (fMRI, PET, EEG)

– Used to predict and explain• Effects of selective brain damage

– E.g. loss of memory consolidation (hippocampus), balance (cerebellum)

– Effects of diseases (e.g. parkinsons)

– Loss of language abilities (hearing = Wernickes; speaking = Broca’s area)

• Operation of emotions (amygdala), and vision (V1-V4)

19

• (in Psychology) Cognitive Behavioural Therapy– Preceded by:

• Psychoanalysis• Behavioural therapy

– Key evidence:• Effectiveness over a short time span• Learning principles (e.g. reward, reinforcement)• Cognitive principles (e.g. selective attention)

– Used to predict and explain• Role of negative/unhelpful cognitions in coping and

anxiety• Importance of retraining in logic, and selective

attention

20

Non-scientific ‘theories’

1. Theories which cannot be tested (cannot be disproven)

– the ‘god’ theory– Psychoanalysis

2. Theories which contain elements which conflict with known facts or established theories (pseudoscience)

– Young earth creationism– Homeopathy

Science does not have a view on untestable theories because there is nothing for it to do! If they cannot be disproven, and new theories built to explain even more in their place, there is no possibility of progress.

Science can be hostile against theories which make claims which contradict well established facts and theories.

21

Avoiding pseudoscience• Psychologists will always be tempted

– We have our own views and intuitions about why people behave in certain ways

– We are as easily mislead as anyone else

• Have scientific knowledge– E.g. Not having any understanding of

electromagnetism leads to a profound vulnerability• Fear of mobile phones, mobile phone towers and power lines

– http://www.skepdic.com/emf.html– http://mybroadband.co.za/news/Wireless/11099.html

• The use of ‘magnetic mattresses’ and bracelets to stop pain– http://www.skepdic.com/magnetic.html

– Deaths result from misunderstanding• Vaccination• Drug use (e.g. alcohol and performance)• Sexually transmitted diseases

22

Avoiding pseudoscience• Understand scientific attitudes and methods

– Never ending scepticism – Ask for replicable evidence– Double blind randomized trials are better than anecdotes

• Confirm that the theories are open to questions and have been evolving– How was the theory discovered?– How is research being used to change/progress understanding

of the theory?• E.g. If the theory was made up out of the blue by a particular person

at a particular time, has never been tested, and has not changed since – it is pseudoscience

• Learn your logical fallacies! (see PSYC1001 manual, and Carl Sagan’s “Baloney Detection Kit” from ‘Demon Haunted World’)

23

Logical errors1. Ad hominem2. Argument from authority3. Appeal to ignorance4. Begging the question5. Observational selection6. Statistics of small numbers7. Inconsistency8. Slippery slope9. Confusion of correlation and

causation10. Straw man11. Weasel words12. The pragmatic fallacy13. Excluded middle

a) Assuming an answer in the way the question is phrased

b) Unwarranted extrapolation of the effects

c) Caricaturing (or stereotyping) a position to make it easier to attack

d) Use of euphemisms and misleading terminology

e) Considering only the two extremes in a continuum of intermediate possibilities

f) Something is true because it works!g) Drawing conclusions from inadequate

sample sizesh) One position contradicts anotheri) Counting the hits and forgetting the

missesj) Attacking the arguer and not the

argumentk) If you are not certain about your

argument, then mine must be truel) Something is true because the person

who said it is of high status.m) Since two things go together, one

must have led to the other.

24

What is science ?

• An attitude – a quest for understanding• No result is ever certain – new evidence is always

welcome

• A technique for discovering the nature of the world

• Anecdote, intuition and blind faith are not the techniques central to science

• Structures of self-correction which (eventually) result in valid knowledge

• If science is wrong, an open forum ensures mistakes are corrected