CHAPTER-2

METHODS
-Gamze Nargileci-
Why we need research?
B15
 Why we need research?

‘I knew they would score better than the other team.’

If someoene becomes convinced that they

accurately predicted an event before it occured is
called hindsight bias.
(''I knew it all along‘’)
▪ trying to understand something withou
t reasoning.
▪ overestimate one's ability to have
foreseen the outcome.
 Psychology is a science and therefore the research needs to be
beased on scientific research which rely on scientific method.

• The process of objectively establishing

facts through testing and experimentation.
• Systematic and planned process, predict
what will happen in advance.
• Interpreting, evaluating variables.

di

THEORY HYPOTHESIS LAW

sets of concepts that statement that describes proving the
explains a a relationship between opinion
phenomena variables, tested
opinions..

prediction
Basic research: expand knowledge and explores
question.research: practical applications, solution based.
Applied

P
‘How human memory works?’
trying to find a solution

‘How we can increase the capacity of the human memory ?’

 Hypothesis express a relation between two variables.

'If I wear my pink t-shirt, I will be very happy.'

ind able d.FIatble

cause s effect
Independent Variable Dependent Variable
The variable that needs to Changes in the study that which
be manipulated to see a
change in the dependent 0
is observed that depends on
the independent variable.
variable.
 ‘Watching violent tv shows makes people more agressive.’

ind able d.FI le

Which show is violent tv show?? What is agression??

Outlining and clearly defining the variables, explaining variables

detaily for measuring process for other researchers to replicate the
original obseervations is called operational definition.
Operational definitions are statements of precise procedures,
it needs to specificy the measurment process and gives
replicability.
details
all
give
 Experiment needs to be valid and reliable in order for it to make
sense and provide you with quality results.

Reliability
7
Outcomes has to be consistent
when the experiment is repeated
Validity

7
Instruments that are used in the
experiment needs to measure
more than once. exactly what you want them to
measure to give accurate results.

IQ test = Week1= 100 score

Week2= 85 score α
Math skills = vocabulary test

not reliable
 Sampling
Selecting individuals to involve in research.

Participants: individuals that participate in research.

Making a research with people that lives in Turkey.

Population:
Individuals that lives
in the same area
Sample: group of people
(Turkey)
selected from population
 Can you choose all the students in a school?

▪ Sample has to represents every individual in the population.

▪ Diversity of the sample is important to give = generalizibility
Using only psychology cannot generalize the findings to
students to make a = all the students in the school.
research

sample error

sampling bias a flawed sampling process that

produces an unrepresentativeness sample.
 Random sampling

▪ Randomly selecting the individuals from the population

With using computer, table of random numbers or flipping coin.

▪ Everyone will have equal chance to be selected
▪ Increase representativeness and diversity.

If we are also including the subgroups as well,

it is called stratified sampling.
 Random sampling increase the generalizability. You can use the
! ‘generalizability’ as the key word for random sampling.
Why? Because it is more representative for the population.
 Describing the Sample
Descriptive Techniques

-Case Study 1
▪ Examines one individual in depth to reveal a full,
detailed picture and get the richest possible
information.

▪ Small group means that the findings cannot

be generalized to a larger population.
(error in sample)

▪ Can be used in brain damage cases or rare

diseases.
 only observing
-Naturalistic Observation
▪ This descriptive method records the beheavior in the natural habitats of the
participants without interacting with them.
Researcher cannot control or change the behavior, they look at a typical
behavior of the participants without manipulating.
Researcher can get a realistic information of the participants’ behavior.

No need for an expensive equipment or a big budget.

 -Ex post facto study
(after the fact, event)

▪ Event already happened

▪ Looking at an effect and seeking for a cause

I
Ex: When researchers found an increase in babies being born with deformed limbs in
a lot of countries in 1970s, they asked the mothers of the babies many questions and
then compared the answers of all of the mothers through correlational analysis. They
found the strongest relationship between the mother taking the drug thalidomide
during the pregnancy and the appearance of the limb deformities in the babies.
 -Meta-analysis

scientific studies. o
▪ a statistical analysis that combines the results of multiple

▪ Meta-analyses can be performed when there are multiple scientific

studies addressing the same question.

Ex: Combining different

experiments about depression
to fully understand what it is.
 -Survey
▪ Descriptive technique for obtaining self-reported attitudes from participants
answers of a series of statements according to their opinions.

▪ Using questionnaries to ask a large number of people questions,

therefore it is not a detailed or in-depth technique.

There is no manipulation, only investigates whether there is a

relationship between the two variables
There is no independent or dependent variable.
Cannot control for many confounding variables.
Psychologists who use surveys in their research have found that
the wording of the questions can affect how people respond. This
is called wording effect or wording bias.

‘’Are you in favor of building a shopping center that could

increase traffic in the area?’’

YES | NO
 “I Have a Friend Who . . .”

The tendency to rely on the anecdotal experiences of a small

number of people is sometimes called the “I have a friend who”
generalization syndrome.
person Anecdotal evidence consists of personal stories about specific
of incidents and experiences. It frequently rely on a single vivid story
opinions rather than solid data which makes it fundamentally flawed.
Ex: A neighbor's experience with their doctor, their review of a
school, or a three-star online rating of a hotel stay.
 Naturalistic observation and surveys can only

o
describe the behavior but cannot explain the
reasons.
They often show one thing is related with another and they correlate.
 Correlations

▪ Explains a relation between variables, if they predict another or not.

▪ Neither variable is manipulated, cannot give cause and effect
relation.

α ‘’Caffeine consumption effects heart problems.’’

Feas y
▪ Correlated variables may influence each other in both directions as
well.
d
‘’Low self-esteem effects depression.’’
r
 What type of relation the variables can have?

Positive Correlation Negative Correlation(inverse relation)

▪ If one variable is present, it ▪ One variable is present,
predicts the other variable. other one is absent.

hours of studying academic success watching tv academic success

 What if there is actually no correlation between the variables?

Illusory correlation is the tendency to perceive a

relationship between infrequent behaviors or traits and
seeing nonexistent infrequent classes of people where none exists. (ex: lucky
relationships items)

A black cat croses Person blames his

a persons path = misfortune on the cat

not relayed
 How strongly two variables can be related?
Ex: Extraversion scores of idenical twins..

To measure the correlation, correlation coefficent(r) will be used.

igtng
Correlation coefficient range from –1 and +1

strong
perfect
negative
-1 0 perfect
positive
+1
INSERT
negative correlation positive correlation
no
correlation

absolute value t not important

 A correlation may be graphed using a scatter plot.
The closer the points come to falling on a straight line, the stronger the correlation.

no correlation

0
aprox.tl
 Experiment
▪ Type of investigation in which a hypothesis is scientifically tested.
▪ A research methods that can give cause and effect relationship.
▪ Manipulates the independent variables to observe changes in the dependent
varibales.
▪ Laboratory
▪ Control andimportant
is very field experiment.
for experiment.
CII
ind variable d variable
''Watching violent tv shows increase agression.''

cause effect
Having an agressive personality
other causes Hungry confounding variables
Sleepy an unmeasured third variable that
influences both the supposed
cause and the supposed effect.
 Ex:''Watching violent tv shows causes agression.''
Sample: 100 participants

related
Experiment group is Control group is
watching something violent. watching a comedy
+ +
show.
take a test to measure aggression take a test to measure aggression

this
mph
▪ Experiments involve two groups. Experimental group and a control group.
no

▪ Assignment is the process by which participans are put into groups.

Students test scores gets affected by distracting sounds in the testing environment.
d ind
▪ If we give participants oppourtunity to choose which group they want to be, this will
create participant-relevant confounding variable but random assignment limits
this.

all subjects have an equal chance of being

assigned to any group or condition in the
study.
▪ Another form is, if the situations of experiment and control group is not equal, this
can create situation-relevant confounding variable.
! Random assignment decrease the confounding variables.
 Other confounding variable is response or participant bias, it
is the tendency for participants to reponse in a certain way.
They usually involve distortions in self-report Data.

00
▪ One form is demand characteristics, they are the clues about the desired or expected
bahavior that researcher can accidentaly give to participant.
Ex: Don’t you think it is bad to steal from someone?

manipulate
answers
▪ Other type is social desirability.
Ex: How many cigarettes do you smoke per day? (respondents may incorrectly choose a lower
number because they feel that’s a more socially accaptable answer.)

▪ If the participants know they are being observed, they can change their behaviours it is
called Hawthorne Effect.

They can be eliminated with using single blind procedure where

participant do not know which group they have been assigned.
▪ Experimenter bias is a special kind of situation-relevant confounding variable. It is the
unconscious tendency for researchers to treat members of the experimntal and control
groups differenty to increase the chance of conforming their hypothesis.

Experimenter bias can be eliminated by using a double-blind procedure. A

mm
double blind occurs when neither the participants nor the researcher are able to
affect the outcome of the research because they don’t know which group
participants are in
 Is it possible to heal with expectations?

▪ A method used is called placebo method refers to

experimental results caused by expectations alone.
▪ Difference in behavior or symptoms caused by
administeration of placebo.

pill with no drug or effect.

▪ Separate the physiological effects of the drug from the
psychological effects of people thinking they took a
drug.
 Pepsi or Coca Cola which one is better ?

All the = pepsi + cola

participants

Participants taste can be affected by the order they drink.

This procedure creates the possibility of order effects. This occurs when
participants' responses in the various conditions are affected by the order of
conditions to which they were exposed.

This can be eliminated by using counterbalancing. With counterbalancing, the

participant sample is divided in half, with one half completing the two conditions
in one order and the other half completing the conditions in the reverse order.

2
1 + cola
Group 1 = pepsi
Group 2 = cola + pepsi
1 2
 deneyimsif
Quasi-Experimental Research

A quasi-experiment is an emprical interventional study used to estimate

the causal impact of an intervention on target population without random
assignment which means it lacks a control element.

Unlike a true experiment, a quasi-experiment does not rely on random

assignment. Instead, subjects are assigned to groups based on non-random
criteria.
Ex: men and women, boys and girls, young and old, or
students in one class and students in another class.

Because of confounding variables, quasi-experiments do not establish cause

and effect relationships, although they can point in the direction of them. This
design is a useful tool in situations where true experiments cannot be used for
ethical or practical reasons.
 Statistics
1-Descriptive Statistics any information
2-Inferential Statistics collected and
measured

1-Descriptive statistics are used to describe, organize, analyze and summarize data.
They provide an overview of numerical data. Key descriptive statistics include; measures
of central tendency, measures of variability, and the coefficient of correlation.

Ex: how many courses does a student

take?
student a: 3

E
student b: 4
student c: 8 Every student's answer
student x: 6 represent a data
student y: 2

{3,4,8,6,2} = frequency distrubition

 Graphs to organize data;

frequency polygon pie chart

bar graph histogram

space in between no space in between
 1.a-Measures of Central Tendency

Attempt to mark the center of the distrubition and finding the typical or average
score. Describe the average or most typical scores for a set of research data or
distribution.of central tendency include the mode, median, and mean.
Measures

6 Number of class a student has? = 3,7,8,4,8

8 3
3 7 8154 ▪ The mean is the arithmetic average of the scores in a
distribution. It is obtained by adding up all the scores 6
and dividing by the total number of scores.
▪ The median is the score that falls exactly in the center of a
distribution of scores.
3 6,708,8 7
▪ The mode is the most frequent score in a distribution.

8
In general, the mean is the most useful measure of central tendency to measure distrubitions.

Normal Distrubition
Normal Curve

4100 mean = median = mode

The mean is sensitive to extreme scores in a distribution, which can
sometimes make the mean misleading. Outlier is an extreme observation or
measurement, that is, a score that significantly differs from all others obtained.
It creates distortions in the nomal graph and cause skeweness as positive or negative.
 Positively skewed

Difficult test mean without the outlier:

gather 10+15+20=45 45/3= 15
Scores = 10,15,20,………..,90
mean with the outlier:
45+90= 135 140/4= 45

mean > median > mode

 Negatively skewed
Easy test
outfier
Scores= 90,85,80,……,3

mean without the outlier:

90+85+80=255 255/3= 85

mean with the outlier:

90+85+80+5= 260 260/4= 65

o's
mode > median > mean
 When a distribution includes outliers,
the median is often used as a better
measure of central tendency. !
normal Mean
outliners median
 1.b-Measures of Variability
Variability refers to how much the scores in a data set vary from each other
and from the mean.
Measures of variability include the range, variance, standard deviation.

a
1.b.x- The range is the distance between the
highest and lowest score in a distribution.

Distrubition= 10,15,20,25,30,35,40 40 10 30
Ex:What is the ranges of ages in your family?
1.b.y- Standart deviation is average distance of any score in the distribution from the
mean.
It indicate the degree to which scores differ from each other and vary around the mean
value for the set. When variability is great, the standard deviation will be relatively large.

mean Yariability

mean
1 1 1 1
20 30 40 50 60 70 80
mean = 50
standart deviation = 10
The taller and narrower frequency polygon shows less variability and
has a lower standard deviation than the short and wider one.
The higher the variance and standard deviation,
the more spread out the distribution.
 Question2: Which has the
highest standart deviation?

a) 30,32, 35, 34
b) 1,5,7,30
c) 5, 10, 13, 18

? ? 60 ? ?
 Scores can be reported in different ways. One example is the standard score or z
score.
▪ Z score tells us how many standart deviation an observation is from the mean.

start calculation here

Z score
paperhanger
+3
-3 -2 -1
keen
0 +1 +2
g
1 standart deviation away from the mean
0 standart deviation away from the mean
▪ A standard score is computed by subtracting the mean
of the distribution from the score of interest and then
dividing the difference by the standard deviation.
Example: Clarence scored 72 on a test, the mean of the
classroom is 80 and standart deviation is 8.
a

64 80 88 96
72
15
 Suppose that we gather data from last years final
chemistry exam and found that it followed a normal
distrubition with a mean of 60 and a standart deviation of
10.
What proportion of students scored less than 50 on an exam?

13 34
151
2 5
30 40 50
60
1.b.z- The standard deviation of the distribution
is the square root of the variance.

σ = standart deviation 22
σ2 = variance
y
 2-Inferential Statistics are used to interpret data and draw
conclusions.
Purpose is to determine whether or not findings can be applied to the larger
population from which the sample was selected.

Making inferences about the generalizibility of the data.

Working with the laws of probability, researchers use inferential statistics to

evaluate the possibility that their results might be due to the fluctuations of
chance.
 Ex: Influence of gender difference in aggression.
How likely chance factor influence the results ?

▪ Many different inferential statistical tests exist such as t-tests, chi square tests, and
ANOVAs. However, what is most important for you to know is that all these tests
yield a p value.

Probability value.
Gives the probability that the difference between the groups is due to
chance. between 0 to 1.
Ranges
0.9 =
0.6 =
0.05 =
“statistically significant
differences were found.”
▪ 0.05 = the cutoff for statistically significant results which
means results are not due to chance.
The smaller the p value, the more significant the results.
A statistically significant result cannot prove that a research hypothesis is correct.
Because p value is based on probabilities, there is always a chance of making an
incorrect conclusion regarding accepting or rejecting the null hypothesis ( H 0 ).

states that there is no difference between

groups or no relationship between variables
if the p value is greater than 0.05, the null
hypothesis is considered to be true.
z

null hypothesis is true : yes, no relationship

null hypothesis is false : no, there is a relationship
 Ethical Guidelines

▪ Psychologists need to act responsibly and morally in their researches

and follow the ethical rules.
makerules
Ethical guidelines established by the APA (American
Psychological Association) for human and animal
research and beaprepared to apply the concepts to
specific research designs.

Any type of academic research must first propose the

apply rules
study to the ethics board or Institutional Review Board
(IRB) at the institution.
reviews research proposals for ethical
violations and/or procedural errors.
 Animal Research
The APA developed strict guidelines about what animals and how
animals can be used in psychological research.

▪ They must have a clear scientific purpose.

▪ The research must answer a specific, important
scientific question.
▪ Animals chosen must be best-suited to answer the
question at hand.
▪ They must care for and house animals in a humane
way.
▪ Animals must be purchased from accredited
companies.
▪ They must design experimental procedures that
employ the least amount of suffering feasible.
 Human Research

▪ No coercion: Participation should be voluntary.

▪ Informed consent: Participants must know that they are involved

in research and give their consent.

Participants should be informed of:

• The purpose, expected duration, and procedures of the
research.
• Their rights to decline to participate and withdraw from the
research once it has begun.
mm
• Factors that might influence their willingness to participate,
such as possible risks, discomfort, or adverse effects.
▪ Anonymity or confidentiality: Participants’ privacy must be protected.
Personal informations of the participants has to be protected.

▪ Risk: Participants should not be exposed to harmful or dangerous research

procedures. This guideline is intended to protect subjects from psychological
as well as physical harm.Participants cannot be placed at significant mental
or physical risk.

α
▪ Debriefing: If an investigation requires some deception of participants
(about matters that do not involve risks), the researcher is required to explain
and correct any misunderstandings as soon as possible. The deception must
be disclosed to subjects in “debriefing” sessions as soon as it is practical to
do so without compromising the goals of the study.