**Research Methods:**

Thinking Critically with Psychological Science

Thinking Critically with Psychological Science

**"Life is lived forwards, but understood backwards"**

- Philosopher Soren Kierkegaard

- Philosopher Soren Kierkegaard

The Scientific Method

In science, everything tends to start with an

observation

; something we notice that makes us pause to think.

Measures of

Central Tendency

In order to summarize and evaluate our data, we must know how to measure it.

Measures of Variation

Random Assignment

Variables

In order to test our hypothesis, we must decide what we are testing for. Experiments examine the effect of

manipulated variables

(known as

independent variables

) on some

measurable behavior or outcome

(known as a

dependent variable

).

**This is also known as**

hindsight bias

: the tendency to believe, after learning an outcome, that one would have

foreseen

it. This is because

"common sense"

more easily

describes what HAS happened than what WILL happen.

hindsight bias

: the tendency to believe, after learning an outcome, that one would have

foreseen

it. This is because

"common sense"

more easily

describes what HAS happened than what WILL happen.

Next, we come up with an

inquiry

(or a question) based on that observation that we can test.

The

hypothesis

is our next step; which is an educated guess or a testable prediction we can make.

(a theory =/= a guess or hypothesis)

Once we know what we're testing, we can

experiment

.

Once we've got our data, we need to

analyze

it to see what our experiment tells us.

Finally, we make a

conclusion

where we either accept or reject our hypothesis based on the data.

The

mean

is the arithmetic average (the sum of all the scores divided by the number of scores)

Just like on a highway, the

median

is in the middle (but you must first organize all the numbers from smallest to largest in order to find it)

The

mode

is the most frequently occurring number (the one that shows up the most)

The

range

of scores is the gap between the highest and lowest score (a crude estimate of variation).

The more useful standard for measuring how much scores deviate is the

standard deviation

(which is a computed measure of how much scores vary around the mean score)

Large numbers of data (heights, weights, intelligence) often form a symmetrical, bell shaped distribution. Most cases fall near the mean, and fewer cases fall near either extreme. we call this type of curve the

normal curve.

The

smaller

the standard deviation (SD), the

skinnier

the curve. The

larger

the SD, the

fatter

the curve.

Confounding variables

are factors that can potentially influence the results of the experiment. We try to control for this by doing

random assignment

when dealing with people or animals.

Constants

(like the name implies) are things we keep the same in the experiment to limit the amount of confounding variables.

No single experiment is conclusive.

Random assignment

is when scientists assign participants to experimental and control groups by

chance.

This minimizes preexisting differences between those assigned to the different groups.

The

participants are also uninformed

about what treatment, if any, they are receiving

. This is an example of an experimental group:

one group receives the treatment, and a contrasting control group does not receive the treatment

or receives a

placebo.

In a

double-blind procedure, neither the participants or the research assistants

collecting the data will

know

what group is receiving treatment (this helps protect against

research bias

).

It's possible

just by thinking

you're getting treatment to boost your spirits, relax your body, and relieve your symptoms. This is known as the

placebo effect

, and is well documented in reducing pain, depression and anxiety.

Observation of Behavior

There are several different ways psychologists observe and describe behavior.

The Case Study:

This is among one of the oldest research methods. Case studies

examine one individual in depth

in hopes of revealing things true of us all. They often suggest directions for further study, and they show us what can happen.

...But individual cases can

mislead

us if the individual being studied is

atypical.

The Survey:

The survey method looks at

many cases in less depth.

Researchers do surveys when wanting to estimate, from a

representative sample

of people, the attitudes or reported behaviors of a

whole population.

...But

asking questions is tricky

, and the answers often depend on the ways questions are worded and the way respondents are chosen.

For instance, people are much more

likely

to approve "

not allowing

" certain things rather than "forbidding" or "censoring" them.

Remember: before accepting survey findings, think critically about the

sample and the sample size

(but you cannot compensate for an unrepresentative sample by simply adding more people)

The Naturalistic Observation:

The naturalistic observation method

records behavior in natural environments

. This can range from watching chimpanzee societies in the wild, to unobtrusively videotaping (and later analyzing) parent-child interactions in different cultures.

Like the case study and survey methods,

naturalistic observations

do not EXPLAIN behavior, they just

DESCRIBE

it. Think of them as snapshots of everyday life that do not control for all the factors that may influence behavior.

Correlation

Describing

behavior is the first step to

predicting

it. Surveys and naturalistic observation often show us that

one trait or behavior is related to another

. In such cases, we say the two

correlate

.

A statistical measure (the

correlation coefficient

) helps us figure how closely two things

vary

together, and thus how well either one

predicts

the other).

Scatterplots

illustrate the

range

of possible correlations, which can go from a perfect positive (+1) to a perfect negative (-1). Each dot represents the

scattered values

of two variables.

A correlation is

positive

if two sets of scores (such as height and weight), tend to

rise or fall together

.

A correlation is

negative

if two sets of scores

relate inversely

(one set going up as the other goes down). An example would be temperature and elevation.

A

weak correlation

, indicating

little relationship

, has a coefficient near zero.

REMEMBER:

Correlation does NOT prove causation!

Although correlation indicates the

possibility

of a cause-effect relationship, it does not prove it.

In addition, when we notice random coincidences, we may forget that they are random and instead see them as correlated. This can result in an

illusory correlation.

Knowing the value of an appropriate measure of central tendency is helpful, but it is also important to know about the

amount of variation

in the data; how similar or diverse the numbers are.

“I walk on x, fly on y”

(I = independent variable, therefore I.V. goes on x axis)

Independent variable starts with “I”, it’s what * I * change

More fun with the placebo effect