RESEARCH DESIGNS:Types of Variables, Confounding variables or extraneous

<< RESEARCH METHODS:To Read Research, To Evaluate Research, To increase marketability

EXPERIMENTAL REASEARCH DESIGNS:Control Groups, Placebo Control Groups >>

Abnormal Psychology PSY404

LESSON 7

RESEARCH DESIGNS

Variable is a noun not an adjective it represents a class of outcomes and it can take more than one value.

Hair color is a variable red, black, white, grey etc.

Types of Variables

Dependent variable (outcome, result, effect, and criterion) is a variable that indicates that whether the

manipulation of the independent variable had an effect.

Independent variable (treatment, factor, and predictor) is a variable that is manipulated to examine its

impact on a dependent variable.

Confounding variable (extraneous, threatening) is a variable that is related to independent or dependent

but not an intended part of study.

Example: vitamin A influences vision of subjects.

Vitamin A

independent variable

Vision

dependent variable

Food

confounding variable

Rich in Vitamin A

Hypothesis

Hypothesis is an educated guess, a research idea:

Hypothesis tells us how two or more variables are related to each other. Research process begins by

generating hypothesis.

Hypothesis Generation

d. By observation Keen observation

e. By studying previous researches.

f. By refuting an old existing theory related to a phenomenon or develops your own theory.

Characteristics of hypotheses

e. Make it testable.

f. Make it supportable.

g. Be sure to have a rational- how theory can help.

h. Demonstrate its relevance how it can solve a practical problem.

When you decide how you want to test your hypothesis you have a research design / plan

d. What will be your sample? How the sample will be selected? What will be the sample size?

e. Instruments / tools / questionnaires to be used

f. Statistical tests to apply

Research design includes aspects that you want to measure in the people ( you are studying that is

dependent variable) and its influence on people's behavior (i.e. independent variable).

Example

Vitamin A influences (or effects) the eye sight of subject.

Vitamin A is independent variable.

Eye Sight is dependent variable.

Research Design

Two groups with equal number of subjects which are equally matched on all characteristics are randomly

selected.

One is experimental group

Abnormal Psychology PSY404

The other is control group

The only difference between experimented and the control group is that of treatment. The experimental

group is exposed or given vitamin A, while the control group does not get any Vitamin A. so if the eye sight

or vision of the experimental group improves or is better than control group then it is attributed to what? It

is because of Vitamin A.

Internal validity is the extent to which a researcher can be confident that the independent variable is

influencing or is causing the dependent variable.

Confounding variables or extraneous

It is any factor occurring in a study that makes the results uninterruptible that pollute or contaminate the

results. If the experimental group was eating diet rich in Vitamin A, or a lot of carrots, that improved their

vision ---. So all these are confounding variables for our research because then independent variable will not

be responsible for bringing changes in dependent variable i.e. The Vitamin A dosage administered to the

experimental group subjects in laboratory will not be the only aspect which will change or influence the

vision or eye sight of the subjects.

Dependent variable / Independent variable

Variables are factors that are connected together in form of hypothesis. The first variable is dependent

variable. Dependent variables are factors or aspects of the study that are influenced by independent variable

and they are expected to change. Independent variable is that aspect that is being manipulated or controlled

and thought to change the dependent variable. In the above mentioned study, Vitamin A is the independent

variable which is varied or given to experimental group but not to the control group. While dependent

variable is eye sight of the subject. So it is Vitamin A which is the independent variable responsible for

bringing change in the dependent variable i.e. eye sight of the subject.

Internal Validity / External Validity

Internal validity is the extent to which results can be attributed to independent variable. A study that is not

open to alternative explanations of the results is set to be internally valid and it is free of confounding

factors or variables. A confounding factor is that, which might have affected the dependent or independent

variable. When a confounding variable is present, the researcher or the investigator cannot know, whether it

is the independent variable or any other confounding variable responsible for the results.

External Validity of a study

In internal validity of the study, we focus only on the results of the study that can be attributed to the

independent variable. While in external validity we want the results to apply to people / samples other than

the subjects of the study or we want to generalize the results to the other settings. That is, if one group of

subjects suffers from depression due to death of a loved one, we want the same results to be applied to

other samples of the population as well. So internal validity and external validity are working in opposition

to one another.

Statistical Significance / Clinical Significance

Statistics is part of psychology, in psychological results; statistical significance means that the probability of

obtaining the observed effects by chance is small. But it is important to understand the difference between

statistical and clinical significance.

Example: Consider a group of adults, who are mentally retarded and they are involved in self injurious

behavior of hitting or slapping themselves. Suppose we are to try a new drug treatment for the self injurious

behavior of adults with mental retardation. We examined one group that receive medication and a second

group that received a placebo (an empty sugar coated pill). To learn whether the new drug diminished or

decreased self injury, we use a rating scale to assess how frequently subjects hit themselves. At the beginning

of the study, all the subjects hit themselves an average of ten times per day. At the end of the study, we

found through the scores on the rating scale that the group on the medication received lower scores or hit

themselves less number of times then the untreated group. So we can conclude that the results are

statistically significant. Statistical significance depends on size of the effect, when you look at the people

Abnormal Psychology PSY404

who were rated as improved. You find that they still hit themselves about six times per day. Although the

frequency is lower. But some of the subjects hit themselves in such a manner that they produce serious cuts

and bruises. This may suggest that your statistically significant results may not be clinically significant i.e.

important to the people who hurt themselves. The effect size that is the actual statistical impact on treated

and un-treated persons in a research can be known by looking at the results of the group as a whole. The

behavioral scientist Wolf (1978) advocated the assessment technique labeled as social validity this technique

involves obtaining information from the person being treated as well as by significant others about the

importance of changes that have occurred. In the example, we might ask the employer, the family members,

friends and others. If they think that the medication has truly reduced the self injurious behavior in the

mentally retarded adults. If the effect of the treatment is large enough to impress those who are directly

involved, the treatment effect is clinically significant. So statistical technique measures the effect size and the

subjective clinical significance measures the social impact on individual or people around him.

The Average Client

Very often we look at results from studies and make generalizations about the group and we ignore

individual differences. The tendency to see all participants as one homogeneous group is labeled as the

patient uniformity myth. Comparing groups according to their mean scores (Group A improved by 50%

over group B). This hides important differences in individual reactions to our interventions and treatments.

The patient uniformity myth leads researchers to make inaccurate generalization about disorders and their

treatment.

It would not be surprising if a researcher studying the example of the treatment of self injurious behavior

concluded that a drug was a good treatment. Although some participants did improve with treatment while

others actually got worse. These fine differences would be averaged out in the analysis of the group as a

whole, the person whose head hitting increased with the new drug, it would make a little difference that this

client and its effect on him will be averaged out. While majority of the people of the same age, gender,

cognitive ability and history of treatment improved, so practitioners who deal with all types of disorders

would consider, that all treatments on a homogeneous group will be the same. The whole sample is treated

as an averaged client.

Studying individual case

The science of abnormal psychology began with careful descriptions of symptoms and arrangement of these

symptoms in to categories.

Descriptive Studies or descriptive approaches are procedures used to summarize an organized sample of

data. They include

1- Observation

2- Case study which focuses on single individual and

3- Survey, which seek to describe a population.

1- Observation includes -----------naturalistic observation

controlled observation

Naturalistic observation is watching animals and humans behave in their natural environment. It gives a

realistic picture of the behavior.

In Participant Observation the observer becomes a participant in the group being observed (to reduce

observer effect).

Observer effect is tendency of animals and humans to behave differently from normal behavior pattern

when they know that they are being observed.

Controlled or laboratory observation is watching animals or humans behavior in laboratory. The advantage

of this observation is that the experimenter has complete control over the situation and specialized

equipment can be used.

2-Case Study

Case study examines and describes in depth an individual's current feeling, thoughts and behaviors.

It investigates intensively one or more individuals, who display the behavior patterns. The case

Abnormal Psychology PSY404

study method relies on a clinician's observations of differences between one person and group with

a disorder, people with other disorders and people with no psychological disorders. The clinician

usually collects as much information as possible, to obtain a detailed description of the person.

Historically, interviewing the person under study yielded a great deal of information on personal,

medical, family background, education, health and work history, as well as the person's own

opinion about the nature and causes of the problems being studied. Case studies are important in

the history of abnormal psychology.

It was S. Freud and E. Kraepelin who used clinical case study. Freud's case studies provided

valuable descriptions of his theory of the development of mental disorders.

Kraepelin's observations of the case studies helped him to construct the first system for the

classification of abnormal behavior.

Wolpe's book psychotherapy by reciprocal inhibition, is based on his treatment procedure called

systematic desensitization, which he applied on some 200 cases, as our knowledge has increased, we

rely less on case study method.

The case study method is valuable in examining rare disorders. Say when a disorder occurs at a very

low frequency, such as 1 case in 10,000, data on such large number of cases is next to impossible.

So we depend on single case.

Mental disorder called Multiple Personality, a rare type of disorder where within one person would

reside two or more then two opposing personalities, a disorder which receives much attention on

the media in films and TV is a target case for single case analysis. All of you have read the famous

fictional characters of Dr. Jekyll and Mr. Hyde (first one a noble character and Mr. Hyde an evil

character). Dr. Jekyll a noble and handsome doctor who had his private laboratory where he use to

go at night and prepare a secret drink, after taking that drink he would become very ugly and

notorious. Then he would go out and kill and rob innocent people. In the morning he would be

noble and kind again.

The famous case of Eve White, a young woman showed Multiple Personality disorder, under stress

Eve White would suffer from a black out period where she would dress like her twin sister Eve

Black, talk like her, would be assertive like her and would not listen to her mother. Hypnosis and

long term psychotherapy gave Eve White the personality of Jane who was sober and assertive. Rare

cases like this one in real life are few.

Drawbacks of case study methods:

Case study lacks internal validity (methodological control)

ii.

Case study lacks external validity ( representative ness)

Advantages of case study:

The case study method is a valuable source for:

Examining rare disorders

ii.

Evaluating and assessing innovative treatment or interventions.

3-Surveys

Surveys provide information about the nature and scope of mental health problems across large

population and regions often leading clues to the causes of disorder. Surveys are an important tool in

epidemiological research which is the study of the incidence and prevalence of disorders in a specific or

specified population.

Incidence refers to the number of new cases of disorder during the specified period of time or incidence

refers the estimated number of new cases during a specific period of time.

Example

Suppose we would like to know the incidence of new cases of college freshmen using drug like Cocaine is

increasing. So prevalence means the overall frequency of the disorder in a specific population.

So the epidemiological data help us to identify the causes of disorder, as in the case of cholera which broke

out in London during the last century. No one knew that how the disease was spreading. But as the data on

incidence and locations of cases were gathered and examined, researcher noticed a pattern and made a

hypothesis, correctly, that the source was contaminated water.

Abnormal Psychology PSY404

In the year 1980, NIMH survey of nearly twenty thousand institutionalized and community

residents revealed that 1 in every 3 US adults had experienced a psychological disorder and that 1 in every 5

was currently experiencing a disorder. The incident of serious psychological disorder is doubly high among

those below the poverty line (Center for Diseases Control) like so many other correlations, the poverty

causes a disorder, raises a chicken and egg question. Does poverty cause disorder? Or disorder cause

poverty? It's both. In schizophrenia a psychotic disorder in which persons looses contact with reality,

understandably leads to poverty. Poverty can also precipitate disorders, especially depression in women and

substance abuse in men.

In case of AIDS virus, epidemiological researchers track the incidence of this disease among several

population (among gay men, among drug users, among spouses and children of infected individuals),

researchers have found how the aids virus is passed from person to person like other types of other co-

relational research, epidemiological research cannot tell us what cause a particular phenomenon but the

knowledge the prevalence and course of disorder tells us and provide understanding in the right direction.

A major concern for survey is external validity if a survey is based on the responses of women from

the posh section of the society, then the results cannot be generalized to the entire population of the city,

because the participants are not representative of the whole population. To ensure representative ness of

the sample, we use random sampling, regardless of the size of the sample, every member of the population

has an equal chance of being included or another method would be to seek participants, who match on the

predetermined picture of the demographic characteristics of the population.

Limitation of Survey method

One obstacle is that not all subjects who agree to participate provide accurate information. Participants may

intentionally distort their answers for a multitude of reason. For example, persons ask about the mental

health of their family members may paint a very bright picture than actually does not exist. A second

obstacle is that the wording of survey questions can influence the answer and detract from the internal

validity of the survey.

Research by Co-relation

When two variables relate to each other, a statistical relationship between the two variables is called a co-

relation.

Example 1

People with depression are more likely to have negative attributions?

One variable is depression and the other one is negative attribution. In order to explore the relationship

between the two, a test or experiment is to be designed to explore it.

Example 2

Exposure of violent films on television is responsible for children's higher level of aggression.

One variable is exposure to television and the other is high level of aggression. Two things are occurring

together. Do not imply that one cause the other i.e. variable X

Causes variable Y cannot be incurred from Co-relation.

In another example allergies and depression may be related. Research reveals a surprisingly high co-relation

between allergic reactions and depressive symptoms. But what causes what. Does depression make you

have allergies?

Does X causes Y or is it that presence of allergies cause people to become more depressed that is Y causes

Or is it that a third variable Z, some common underline biological factor which make people become

allergic and depressed. Is it that Z causes X and Y?

Kinds of Co-relation

There are two types of co-relation, Positive co-relation and negative co-relation. Increase in one variable is

associated with increase in same strength and quantity in the other variable, it is called positive correlation.

Decrease in one variable is related with decrease in same strength and quantity of the other is also labeled as

Abnormal Psychology PSY404

positive co-relation. Perfect positive co-relation is +1.00. Negative Correlation is when increase in one

variable leads to decrease in another variable. It is negative correlation. Perfect negative correlation is -1.00.

The correlation coefficient is represented by small r. so correlation allows us to see whether a relationship

exists between two variables but it does not conclude either variable cause the other. So it means X is

related with Y but we do not know whether X causes Y?

Or Y causes X? Or a third variable Z causes variable X and Y?

Epidemiological Research

Epidemiology is the study of the incidence, distributions and consequences of a particular problem or set of

problems in one or more population. One strategy is to determine the incidence of disorder, the estimated

number of new cases, during a specific period of time.

A related strategy is involved determining prevalence, the number of people with a disorder at any one time.

Epidemiologists study the incidence and prevalence of disorders among different groups of people. The

epidemiology research aims to determine the medical problems and psychological problems. In 1900, a

number of Americans displayed symptoms of strange mental disorder. Its symptoms were similar to organic

psychosis, which is caused by taking drugs or great quantities of alcohol. Most of the victims were poor and

they were African Americans. Using the method of epidemiological research Gold Berger found

correlations between the diet and the disorder, and he identified the cause of disorder as a deficiency of the

Vitamin B, their diet was improved and the symptoms were eliminated. A long term wide spread benefit of

Gold Berger findings was introduction of Vitamin enriched bread in the 1940.

In 1980, Mount St. Helens volcano erupted, creating extensive property damage and loss of life. When

subjects of a comparable (control) community that was similar in demographics but had not experienced a

similar traumatic event, the researchers found a significant high number of psychological disorders in the

people who live near Mount St. Helens. This is a co-relational study. There is a relationship between stress

and psychological problems.

Table of Contents: