RE󰈠󰉋󰉚󰈣C󰉀

ME󰈜󰈿󰈮D󰈭󰈴O󰉁Y
MO󰉍󰈖󰈳󰉈 I󰈾󰈽: RE󰈠󰉋󰉚󰈣C󰉀 󰉌ES󰈾󰉂󰈯

SA󰈖󰈲󰈄󰉚 D󰉀I󰈱󰉝N
VI󰈠󰈾󰈙󰈽N󰉂 󰉆AC󰈖󰈴󰈙Y
A󰈾B󰉚󰈠
 RE󰈠󰉋󰉚󰈣C󰉀 󰉌ES󰈾󰉂󰈯
“As a process of creating
“A traditional research

01 02
an empirical test to design is a blueprint or
support or refute a detailed plan for how a
knowledge claim” research study is to be
(Mertens, 2010) completed—operationalizi
ng variables so they can be

03
“A research design is a measured, selecting a
plan, structure and sample of interest to
strategy of investigation study, collecting data to be
so conceived as to obtain used as a basis for testing
answers to research hypotheses, and analysing
questions or problems” the results.” (Thyer, 1993)
(Kumar, 2011)
➔ View it like a glue that holds various components of a research together.
➔ In simple words, blueprint.
➔ It is important because: it answers your research question validly, objectively,
accurately and economically.
➔ Decisions regarding what, where, when, how much, by what means
concerning an inquiry or a research study constitute a research design (Kothari,
2004).
➔ It shows how different components of your research: 1) —the sample, 2)
measures, 3) treatments or programs, and 4) methods of assignment come
together to answer the hypothesis.
➔ Functions of your research design
(Kumar, 2011):
1) conceptualise an operational
plan to undertake the various
procedures and tasks required
to complete your study.
2) ensure that these procedures
are adequate to obtain valid,
objective and accurate answers
to the research questions.
➔ A good research design= A good study! Helps to maintain the quality of your
results.
➔ It must be practical.
➔ A particular design is suitable for a particular problem.
➔ Remember, it is systematic plan.
 IM󰈪󰈮󰈣T󰉚󰈰󰉎E 󰈮F 󰈤󰉈󰈟E󰉝R󰉑󰈿 D󰉈󰈠I󰉁N
➔ It reduces inaccuracy;
➔ Helps to get maximum efficiency and reliability;
➔ Eliminates bias and marginal errors;
➔ Minimizes wastage of time;
➔ Helpful for collecting research materials;
➔ Helpful for testing of hypothesis;
➔ Gives an idea regarding the type of resources required in terms of money,
manpower, time, and efforts;
➔ Provides an overview to other experts;
➔ Guides the research in the right direction.
 FE󰉝󰈜󰈓󰈣ES 󰈮󰉇 󰉚 󰉁O󰈮D 󰈤󰉈󰈟E󰉝R󰉑󰈿 D󰉈󰈠I󰉁N

1) Objectivity
2) Reliability
3) Validity
4) Generalizability
Some other considerations: 1) Feasibility 2) Flexibility
3) Free from confounding variables 4) Minimize
biases.
 T󰇳󰇴ES 󰈮󰉇 󰈣󰉈SE󰉝󰈤󰉎H (󰈲ar󰇹󰉜y󰈔, De󰈲󰈀󰉅󰇵o &
Fes󰉃󰈎󰈞g󰇵󰈸, 2005)
Ex󰈥e󰈹󰈎m󰇵󰈝󰉄al Qu󰈀s󰈏-E󰉕󰈦er󰈎󰈚󰇵󰈞ta󰈗 Non-󰈩󰉕󰈦󰇵ri󰈚󰈩󰈞t󰇽󰈗

The gold standard of Quasi= Sort of Weakest design.

research: Most rigorous. Often used in medicine or No IV Manipulation.
(Maxwell, 2004) neuroscience studies. Random Assignment not
Cause-effect. Lacks random possible.
Manipulation of IV. assignment/control group. E.g.: Smoking
IV
Manipulation/Quasi-independent
ID󰉋󰈰󰈙󰈽FI󰉑󰉝󰈙󰈽ON 󰈮󰉇 󰉌󰉈SI󰉂󰈯
 EX󰈪󰉋󰈣󰈽ME󰈰󰈙󰉝L 󰉍󰉈󰈟IG󰈰
➢ First given by R.A. Fisher.
➢ “An experiment usually consists in making an event occur under known
conditions whereas many extraneous influences as far as possible are
eliminated and close observation is possible so that relationship between
phenomena can be revealed.” (Beveridge, na).
➢ Experimental designs are used to prove cause and effect.
➢ They are rigorous and minimized internal validity threats.
➢ The researcher’s objective is to show that the treatment or intervention
produced an effect and that this effect could not possibly have been due
to anything besides the treatment.
 EX󰈪󰉋󰈣󰈽ME󰈰󰈙󰉝L 󰉍󰉈󰈟IG󰈰
➢ Keep in mind that in some situations experimental designs either cannot be used
or would be unethical
➢ Terms like conditions, hypotheses and manipulation are central to it.
➢ It might also be referred as randomized control trial.
➢ A randomized control trial has 3 principles given by Fisher: 1) Replication 2)
Randomization 3) Local control.
➢ Replication: It is done to increase the accuracy of the results. Deliberate
repetition of the experiment. Nearly identical experiment procedure, different
participants, different time (Singh, 2017). This is the reason we operationalize our
variables.
➢ Randomization: Allocation of treatment to experimental units at random. It
protects against extraneous variables.
 EX󰈪󰉋󰈣󰈽ME󰈰󰈙󰉝L 󰉍󰉈󰈟IG󰈰
➢ Local control: Extraneous factors can introduce variability. The aim
here is to make blocks where homogeneity is maintained and reduce
experimental error. Holding extraneous variables constant.
➢ Experimentation can be considered a technique of deliberately staging a
situation designed to force nature to provide a “yes” or “no” answer to a
specific hypothesis concerning the phenomena under discussion.
➢ 1) Randomization 2) Control Group 3) Intervention
 EX󰈪󰉋󰈣󰈽ME󰈰󰈙󰉝L 󰉍󰉈󰈟IG󰈰: 󰉎HA󰈤󰉝󰉎T󰉈󰈤I󰈟T󰈾󰉑󰈟
➢ Manipulation of one or more IV: This means at least 2 levels have to
there. Each level is referred as treatment/condition. Remember
conditions in control and experimental group are the same except the IV
exposure.
➢ Control over extraneous variables
➢ Randomization
 Introduction
The Progressive Era was in the late nineteenth and
early twentieth centuries during a period where
there were several social, economic and political
reforms to end the dominance of big business and
increase equity
 Ad󰉏a󰈞t󰈀󰈇󰇵󰈼 of Dis󰈀󰇷󰉐󰇽n󰉃a󰈈󰈩s 󰈢󰇾
Ex󰈥e󰈹󰈎m󰇵󰈝󰉄al 󰇷󰈩󰈼󰈏g󰈝 Ex󰈥e󰈹󰈎m󰇵󰈝󰉄al 󰇷󰈩󰈼󰈏g󰈝

1. Ability to identify 1. Cannot be used if IV is

causal relation. something you cannot
2. Low internal and manipulate.
external threats. 2. High control on extraneous
3. Minimized biases and variables make it difficult to
errors be applied outside laboratory
conditions.
 Bet󰉒󰈩󰇵󰈞 Gro󰉉󰈥 󰉍󰇵󰈼ig󰈝/In󰇷e󰈦󰈩n󰇷󰇵󰈞t
Me󰈀s󰉊󰈸e󰈼/Un󰈸e󰈘󰈀t󰇵󰇷 Sa󰈛p󰈗󰈩
➔ Random assignment to 2 groups of your participants.
➔ This means 1 participant exposed to one condition only.
➔ Compared two different individuals.
➔ Differences between the two groups measured.
➔ Control extraneous variables.
➔ E.g. If your IV is caffeine intake. You can expose different groups to different levels
of caffeine and then subsequently measure their say puzzle solving ability or test
scores.
➔ Independent scores.
➔ For a between-subjects experimental design, individual differences are a particular
concern and can create serious problems
➔ Assignment bias.
 Ad󰉏a󰈞t󰈀󰈇󰇵󰈼 of B󰈩󰉃󰉓󰇵en-󰈇󰈹󰈡󰉊p 󰇷e󰈼󰈎g󰈝

1. Simplest design to carry out.

2. Because each participant is measured only once, the researcher can be reasonably
confident that the resulting measurement is relatively clean and
uncontaminated by other treatment factors.
3. No order effects
4. Since exposure is only to one condition: no threat of practice or experience.
5. No fatigue or boredom.
6. contrast effects that result from comparing one treatment to another (a
60-degree room might feel cold after a 70-degree room, but the same 60-degree
room might feel warm after a 50-degree room).
 Dis󰈀󰇷󰉐󰇽n󰉃a󰈈󰈩s 󰈢󰇾 Be󰉄w󰈩󰇵󰈝-󰈈ro󰉉󰈥 󰇶󰇵si󰈇󰈞

1. Requires larger sample size: This can be a problem for research involving special
populations in which the number of potential participants is relatively small.
2. Cannot learn about the nature of relationship.
3. Individual differences: each score is obtained from a unique individual who has
personal characteristics that are different from all of the other participants. The
concern with individual difference is that they can provide an explanation for
why two different individuals produce two different scores in a research study:
i.e. confounding variables.
 Wit󰈊󰈎󰈞 Gr󰈢u󰈥 󰉍󰈩󰈼󰈏g󰈝/Rep󰈩󰇽󰉃e󰇶 M󰈩󰇽su󰈸󰈩󰈼/Rel󰈀󰉃󰇵󰇶
Sam󰈥󰈘󰈩
➔ In this design, each participant is tested under all conditions. Therefore, in such
a design there will only be one group.
➔ Repeats measurements of the same individuals under different conditions.
➔ Differences between conditions measured.
➔ Overcomes with the help of counterbalancing- different participants will be
tested in different order of conditions. For example, ABC, ACB, BAC, BCA,
CAB, CBA order of conditions A, B and C
 Ad󰉏a󰈞t󰈀󰈇󰇵󰈼 of W󰈎󰉃󰈋󰈏n 󰈇󰈹o󰉉p 󰇷󰇵󰈼ig󰈝

1. Fewer Participants. For example, to compare three different treatment

conditions with 30 participants in each treatment, a between-subjects design
requires a total of 90 participants (three separate groups with 30 participants in
each). Saves time and money.
2. Useful when participants are difficult to find.
3. Eliminates problems arising due to individual differences.
4. Performance differences across treatments cannot be due to error variance
arising from such differences: we can separate the individual differences from
the rest of the variance in a within-subjects design.
 Dis󰈀󰇷󰉐󰇽n󰉃a󰈈󰈩s 󰈢󰇾 Wi󰉄h󰈎󰈝 󰈈r󰈢u󰈥 󰇶󰈩s󰈏󰈇󰈞

1. Since all subjects are exposed to all the conditions or treatments: can be
demanding.
2. Recruiting participants can be difficult.
3. Boredom + fatigue
4. May fail to show up for future sessions if your strategy is to divide these
conditions over days.
5. Carryover effects
6. Order effects
7. Demand characteristics.
 Mat󰇹󰈋󰈩d G󰈸󰈢u󰈦 󰉍󰈩s󰈏󰈇󰈞
➔ Participants are prescreened and then assigned to each condition. Same amount
of high and low scores in each group.
➔ We saw in between-group design that researchers may use matching technique
to avoid assignment bias.
➔ It is a variation of that.
➔ Instead of hoping that random assignment produces equivalent groups, a
researcher can use matching to guarantee that the different groups of
participants are equivalent (or nearly equivalent).
Overcoming the Individual differences
➔ Randomization: Fair and Unbiased. But it is not perfect.
➔ Matching: Identify variables that can influence results. Matching
involves assigning individuals to groups so that a specific variable
is balanced, or matched, across the groups. The intent is to create
groups that are equivalent (or nearly equivalent) with respect to the
variable matched.

➔ Holding variables constant: eliminate it.

 Ad󰉏a󰈞t󰈀󰈇󰇵󰈼 of M󰈀󰉃󰇸h󰇵󰇷 󰈈ro󰉉󰈥 󰇶󰇵si󰈇󰈞

1. Allows you to control subject variables that may otherwise obscure the effect of
the independent variable under investigation.
2. You may be able to demonstrate a given effect with fewer subjects, thus saving
time and money
 Dis󰈀󰇷󰉐󰇽n󰉃a󰈈󰈩s 󰈢󰇾 Ma󰉄c󰈊󰈩󰇶 g󰈸󰈢u󰈦 d󰈩󰈻󰈏󰈈n

1. Matched designs require you to use somewhat modified versions of the

inferential statistics you would use in an unmatched, completely randomized
design. These statistics for matched groups are somewhat less powerful than
their unmatched equivalents.
2. matched characteristic has little or no effect on the dependent variable, then
matching will do no good.
3. When using a matched design, you also must be sure that the instrument used
to determine the match is valid and reliable.
 NO󰈰-󰉋󰈅P󰉈󰈤I󰈱󰉋N󰈜󰉚󰈳 DE󰈠󰈾󰉁N

A research that lacks: 1) Manipulation 2) Random assignment of participants.

Does not aim at reducing or controlling confounding variables.
The non-manipulated variable is usually a participant characteristic (such as male versus
female).
Often used when it is unethical to assign participants to different conditions.
Researcher here considers other possible explanations.
Interestingly, no consensus in classification of non-experimental designs ( Johnson, 2001).
 NO󰈰-󰉋󰈅P󰉈󰈤I󰈱󰉋N󰈜󰉚󰈳 DE󰈠󰈾󰉁N
The most frequently used labels in these texts were survey (twelve times), correlational (ten
times), descriptive (eight times), and causal-comparative (five times).
Again, the choice between the experimental and nonexperimental approaches is generally
dictated by the nature of the research question.
Retrospective in nature.
Survey and correlational designs are descriptive non-experimental research.
What do you think about punishment in classrooms?
How do you feel about drug use by students?
What about euthenesia?
Suppose we hypothesize that texting while
driving is more prevalent among younger
age groups, as has been tested in the
published literature (Harrison, 2011; Ling,
Bertel, & Sundsøy, 2012). We could use a
survey research design by asking a sample
of young college students who drive to
indicate in a questionnaire how often they
use text messaging while driving (per
month).
 Sur󰉏󰈩󰉙 R󰇵se󰈀󰈸󰇸h
➔ The survey research design is the use of a survey, administered either in written
form or orally, to quantify, describe, or characterize an individual or a group.
➔ Survey research is used to assess people’s thoughts, opinions, and feelings.
➔ Idea: You want to know what people think about something: ask them!
➔ Therefore, a survey can also be called a questionnaire or self-report.
➔ Surveys can provide us accurate, reliable, and valid data, but to do this they
require serious effort and thought.
➔ Most surveys ask a large number of people (usually called respondents) about
their beliefs, opinions, characteristics, and past or present behaviors.
➔ Data can be in quant or qual form.
➔ Characteristic: 1) Involves selecting a sample
2) Predetermined set of questions.
➔ We usually sample many respondents and
ask all of them the same questions.
➔ We measure many variables with the
questions and test multiple hypotheses
simultaneously.
➔ Social desirability.
➔ Can be reliable only if: representative
sample.
➔ Conclusions drawn from survey research
can be erroneous otherwise.
➔ Survey Methods: Interviews, Telephonic interviews, Electronic surveys, Opinion polls,
Questionnaires (Written).
➔ Read upon them in detail: advantages/disadvantages for your exam.
Ite󰈚󰈼 󰈎n S󰉊󰈸󰉐ey D󰈩󰈻󰈏󰈈n󰈻:
 CLOSE ENDED OPEN ENDED
1. A restricted item, also called a 1. An open-ended item is a question or
closed-ended item, is a question or statement that is left completely
statement in a survey that includes “open” for response.
a restricted number of answer 2. More than yes/no
options to which participants must 3. “How do you think we should
respond. finance public education?”
2. Likert scale 4. They can also increase the
3. survey responses can be easily respondent’s sense of control while
entered or coded for the purposes filling out the survey.
of statistical analysis. 5. Open-ended items are most often
4. main limitation of using restricted used with the qualitative research
items is that the analysis is design because the responses in the
restricted to the finite number of survey are purely descriptive
options provided to participants
Suppose we hypothesize that texting while
driving is more prevalent among younger age
groups, as has been tested in the published
literature (Harrison, 2011; Ling, Bertel, &
Sundsøy, 2012). So, we ask a sample of
participants who drive to indicate in a
questionnaire their age (in years) and how
often they use text messaging while driving
(per month).
 Cor󰈸󰈩󰈘󰇽ti󰈡󰈝󰇽󰈘 Res󰈩󰇽󰈸󰇸h
➔ Often used in pilot testing.
➔ Controlling other possible factors: X
➔ Linear relationship between variables: Variables can be either positively, negatively
or not related.
➔ A statistical procedure called correlational analysis is used to ascertain the extent
of the relationship among individual scores on the two variables (tests)
➔ When is correlational research preferred? 1) Statistical relation is not a causal one
2) independent variable is impossible, impractical, or unethical to manipulate.
➔ Variable need not be always quantitative in nature!
Early civil rights movements

Martin He was an important leader of the

L. King civil rights movement

Rosa Refused to give up his seat on a

Parks public bus to a white person

John Civil rights leader and politician who

Lewis helped plan the March on Washington
➔ Positive: Variables vary in same direction.
➔ Negative: Variables vary in opposite direction.
➔ The correlation coefficient is a statistic used to measure the strength and
direction of the linear relationship, or correlation, between two factors. The
value of r can range from −1.0 to +1.0.

HW: Read up on advantages and disadvantages of a correlational research.