COURSE: RESEARCH METHOD-II

RESEARCH DESIGN
What is research design?
A research design is the plan or framework used to conduct a research study. It involves
outlining the overall approach and methods that will be used to collect and analyze data in order
to answer research questions or test hypotheses.
Purpose of research design?
 To provide a plan in answering research questioning.
 Each design has its own applicability depending on problem and objective of study.
Importance:

 Clear direction: the main role is providing direction to the research and guiding the choice of
clear objectives, which help the researcher to focus on the specific research questions or
hypotheses they want to investigate.

 Control: researchers can control variables, identify potential confounding factors, and use
randomization to minimize bias and increase the reliability of their findings.

 Replication: Research designs provide the opportunity for replication. This helps to confirm the
findings of a study and ensures that the results are not due to chance or other factors. Thus, a
well-chosen research design also eliminates bias and errors.

 Reliability: Benefits of research design also include reducing inaccuracies and ensuring the
reliability of the research (i.e., consistency of the research results over time, across different
samples, and under different conditions).

 Efficiency: A strong research design helps increase the efficiency of the research process.
Researchers can use a variety of designs to investigate their research questions, choose the most
appropriate research design for their study, and use statistical analysis to make the most of their
data. By effectively describing the data necessary for an adequate test of the hypotheses and
explaining how such data will be obtained, research design saves a researcher’s time.

Validity: A research design ensures the validity of the research, i.e., whether the results truly
reflect the phenomenon being investigated.

Threats to validity:

 History: it refers to the events that may occur during the time frame of the study which
are not actually a part of study. But they produce effects that influence the results of
study either increasing or decreasing the expected results.
  Selection: it occurs when respondents of the study ae chosen not only individual but as a
group.
 Testing: it refers to the pre-test given that results in an improved performance in the
post-test. To avoid this threat pre-test may not be need to administration. However, if the
pre-test is given, another measure is recommended to use as an alternate form of
instruments.
 Instrumentation: it refers to unreliability in measuring instruments that may result to an
invalid measurement of performance. The change in instruments used between the pre-
test and the post-test may result in an effect not caused by a treatment introduced.
 Maturation: this factor refers to the physiological and psychological changes that may
happens to respondent of the study over a period of time. If the time frame of training
program is quite long and rigid the participant may experience some psychological
discomfort due to boredom, tiredness, hunger and like that.
 Mortality: it refers to the loss of participants during the post-test stage or even during the
implementation of the study. When the same group of individuals is studied over a long
period of time. By the time a follow-up study is conducted on the same group, some
member may have dropped out or may refuse to corporate future in study.

Different types of research design

Broadly, research design types can be divided into qualitative and quantitative research.

Qualitative research is subjective and exploratory. It determines relationships between collected

data and observations. It is usually carried out through interviews with open-ended questions,
observations that are described in words, etc.

Quantitative research is objective and employs statistical approaches. It establishes the cause-
and-effect relationship among variables using different statistical and computational methods.
This type of research is usually done using surveys and experiments.

Qualitative research vs. Quantitative research

Qualitative research Quantitative research

Deals with subjective aspects, e.g., experiences, Measures different types of variables and describes
beliefs, perspectives, and concepts. frequencies, averages, correlations, etc.

Tests hypotheses about relationships between

Deals with non-numerical data, such as words,
variables. Results are presented numerically and
images, and observations.
statistically.
 In qualitative research design, data are collected via
direct observations, interviews, focus groups, and Quantitative research design is empirical. Data
naturally occurring data. Methods for conducting collection methods involved are experiments,
qualitative research are grounded theory, thematic surveys, and observations expressed in numbers.
analysis, and discourse analysis. The research design categories under this are
descriptive, experimental, correlational, diagnostic,
and explanatory.

Data analysis involves interpretation and narrative Data analysis involves statistical analysis and
analysis. hypothesis testing.

The reasoning used to synthesize data is inductive. The reasoning used to synthesize data is deductive.

Typically used in fields such as sociology, Typically used in fields such as economics, ecology,
linguistics, and anthropology. statistics, and medicine.

Example: Focus group discussions with women

farmers about climate change perception. Example: Testing the effectiveness of a new
treatment for insomnia.

QUANTATIVE RESEARCH DESIGN

Survey designs: are procedures in quantitate research in which researcher administer a survey to
a sample or to the entire population of people to describe the attitudes, opinion, behaviors, or
characteristics of population. In this procedure, survey researchers collect quantitively, numbers
data using questionnaires (e.g. mailed questionnaires) or interviews and statistically analyze the
data to describe trend about responses to questions and to test research questions or hypothesis.
When do we use survey research?
 to describe trends such as community interest in school bond issues or state or national
trends about mandatory student policies?
 To determine individual opinions about policy, issues, such as whether students need a
choice of school to attend
 Survey help identify important beliefs and attitudes of individuals, such as college
students’ beliefs about what constitutes abusive behaviors in dating relationships
 Survey as provide useful information to evaluate programs in schools such as the success
of a robotics program in science education.
Nature, goals: The goal of a survey is to collect data from a representative sample of a
population to draw conclusions about that larger population
basic step in survey research design:
step1: decide if a survey is the best design to use; you need to decide whether survey research
is the best design to use in the study. Surveys help to describe the trends in population or
describe the relationship among variables or compare groups. Examples where surveys are most
suitable are to access trends or characteristics of a population; learn about individuals attitudes,
beliefs and practices; evaluate the success or effectiveness of a program; or identify the needs of
community.
Step 2: identify the research questions or hypothesis: You can address both research
questions and hypotheses in a survey design. Surveys lend themselves to hypothesis testing
because you will be studying a sample to draw inferences to a population. Forms of research
questions or hypotheses are those that:
Describe the characteristics or trends of a population of people, such as the frequency of tobacco
use among male high school students
◆ Compare groups in terms of specific attributes, such as a comparison of teachers and
administrators about attitudes toward "in-service" learning days
◆ Relate two or more variables, such as a survey of teachers to relate "burnout" to number of
years of teaching.
Step 3. Identify the Population, the Sampling Frame, and the Sample:The process of survey
research begins with identifying the population. This step requires defining the population,
determining the number of people in it, and assessing whether you can obtain a list of names
(i.e., the sampling frame) for the sample. Also, the population may need to be stratified before
sampling, so select characteristics of the population (e.g., males and females) are represented in
the sample.
Once you have identified the target population and compiled a list of its members, you can select
the sample, preferably using random sampling procedures. You will need to identify an adequate
sample size, using a sampling error formula.
Step 4. Determine the Survey Design and Data Collection Procedures: The researcher must
also determine if the survey study will be cross-sectional or longitudinal.. The decision to use a
longitudinal or cross-sectional design relates to the nature of the problem studied, access to
participants, and the time available to the researchers for data collection.
Step 5. Develop or Locate an Instrument: You need an instrument to collect or measure the
variables in a study. It is easier to locate an instrument than to develop one. Standards of
reliability and construct validity need to be applied to scores from existing instruments before
you select them for use. If a study addresses only a few variables, researchers can design their
own instruments. A check for the reliability and validity of scores from this instrument during
data analysis is most important.
Step 6. Administer the Instrument: This step is perhaps the most time-consuming phase of
survey research. It involves seeking and obtaining permission to conduct the survey and using
procedures for data gathering, such as training interviewers or preparing questionnaires for
mailing. It requires continually following up to obtain a high response rate, checking for
response bias if questionnaires are used, and preparing the data for analysis by coding the
information from the instruments into a computer file.
Step 7. Analyze the Data to Address the Research Questions or Hypotheses: The data
analysis procedures will reflect the types of research questions or hypotheses the researcher plans
to address in the study. Analysis consists of noting response rates, checking for response bias,
conducting descriptive analysis of all items, and then answering descriptive questions. It might
also involve testing hypotheses or research questions using inferential statistics.
Step 8. Write the Report: You should write the survey study using a standard quantitative
structure that consists of an introduction, the review of the literature, the methods, the results,
and the discussion. Specify in the "Methods" section of the study detailed information about the
survey procedures. Include in the "Discussion" section comments about the generalizability of
the results to the population.
Survey methods
Written survey: Written administration of questionnaires can take several
forms: The questionnaires may be administered to a group, they may be
dropped off at a particular location, or they may be mailed to the
respondents. Group administration is familiar to you if you have completed
course evaluations in college. Group administration is a very efficient use of
time and money and can have a very high response rate if attendance by
group members is high, as in a class. Drop-off administration is often done
by an organization, such as a church, that has many members who attend a
particular location over some period of time but may not all be present at
one time. Copies of the questionnaire are dropped off in a particular location
to be picked up at some later time. Mail administration is familiar to all.
Mail surveys: mail surveys are quick and convenient. they are used to
distribute self-administered questionnaires that respondents fill out on their
own.
 Advantages: they usually can be completed relatively quickly.
Because they are self-administered. Mail surveys also avoid the
problem due to interviewer bias. It is best method for dealing with
highly personal or embarrassing topic, especially when anonymity of
respondents preserved.
 Disadvantages: cost of copying and mailing. Respondents will not be
able to ask questions. Researcher has little control over order in which
respondent answer the questions. Oder of questions may affect how
respondents answer the questions. Mail survey is low response rate
 that can result in response rate bias. Response rate refers to
percentage of people who complete the survey. For example, 30 out
100 people complete the survey the response rate is 30%. A low
response rate indicates that there could be a response rate bias that
threats the representativeness of a sample. For example, respondents
with literacy problems low education level vision problem may not
complete the survey. Often people randomly selected for a sample are
too busy or not interested enough in the study to return a completed
questionnaire.
Personal interviews: face to face conversation is called personal interview.
it is used to collect survey data in which respondents are usually contacted
in their homes or in a shopping mall.
 Advantages: Personal interview allows greater flexibility in asking
questions than does the mail survey. During interview the respondent
can obtain clarification when question is unclear and trained
interviewer can follow up incomplete or ambiguous answer to open
ended questions. Interviewer control the order of questions and ensure
that all respondents complete the questions in same order. Response
rate of personal interviews has been higher than mail survey
 Disadvantages: personal interview is costly. The use of trained
interviewers is expansive in terms of both money and cost. Another
disadvantage is interviewer bias. The interviewer should be neutral
medium through which questions and answers are transmitted.
Interviewer bias occur when the interviewer records only selected
portion of respondents answer or tries to adjust wording of a question
to fit the respondent. To reduce interviewer bias research can employ
highly motivated, well-paid interviewers who are trained to follow
question wording exactly, to record responses accurately, and to use
follow-up questions judiciously. Interviewers should be closely
supervised by director of survey project.
Telephonic interviews: it is used for brief surveys.
 Advantages: It also provides better access to dangerous
neighborhood locked buildings and respondents available only during
evening hours. Interviews can be completed more quickly when
contact are made by phone and interviewers can be better supervised
when all interviews are conducted from one location.
 Disadvantages: selection bias exist when respondents are limited to
those who have telephones and problem of interviewer bias remains.
There is a limit to how long respondents are willing to stay on the
phone and individuals may respond differently when talking to faceless
voice than they would to a personal interviewer. Individuals from high
 socioeconomic group are more likely to have multiple phone numbers
and hence might be overrepresented in a survey based on random-
digit dialing. People may forget response option offered by interviewer.
Internet survey: computer-based survey. it is easy to administer
participants complete a questionnaire online and click on submit button to
have their responses recorded. There is multiple software that help to record
and summarize the responses.
 advantages: This method is efficient and millions of participants who
vary in age, ethnicity and nationality can be contacted through a few
keystrokes on a computer. Online questionnaires are paperless thus
saving natural resources and copying costs. Participants may respond
when it is convenient and do so without leaving the comfort of their
home, office room.
 Disadvantages: potential for sample biases at present, there is no
way to generate a random sample of internet users. Selection bias is
present because sample consist of individuals who have internet
access. Higher income group have more internet access. Lack of
control over research environment is also major disadvantage.
Because researcher is not present there is no easy way to determine if
respondents have a clear understanding of instructions.
Survey research design types
The three types of survey design are the cross-sectional design, the successive independent
samples design, and the longitudinal design.
Cross-sectional design: In the cross-sectional design, one or more samples are drawn from the
population(s) at one time. Cross-sectional designs allow researchers to describe the
characteristics of a population or the differences between two or more populations, and
correlational findings from cross-sectional designs allow researchers to make predictions. In a
cross-sectional survey design, the researcher collects data at one point in time. For example,
when middle school children complete a survey about teasing, they are recording data about their
present views. For example, a cross-sectional design was used in a nationwide study of Internet
use among 1,100 teens aged 12–17 This design has the advantage of measuring current attitudes
or practices.
Advantages: It also provides information in a short amount of time, such as the time required for
administering the survey and collecting the information. Cross-sectional designs are ideally
suited for the descriptive and predictive goals of survey research.
Successive independent samples design: A successive-independent-samples design draws
multiple random samples from a population at one or more times. different samples of
respondents from the population complete the survey over a time period. The successive
independent samples design allows researchers to study changes in a population over time.
For successive independent samples designs to be effective,
 the samples must be drawn from the same population, and must be equally representative
of it. If the samples are not comparable, the changes between samples may be due to
demographic characteristics rather than time.
 In addition, the questions must be asked in the same way so that responses can be
compared directly. this design measures the differences in individual participants’
responses over time.
For example, public opinion researchers frequently ask independent samples of Americans the
extent to which they approve of the U.S. president (referred to as the president’s “approval
ratings”). Changes in approval ratings over time are used to characterize Americans’ opinions of
the president’s actions
Limitation:
 The purpose of the successive independent samples design is to describe changes over
time in the distribution of population characteristics, not to describe changes in individual
respondents. Accordingly, the successive independent samples design is not always
helpful in ferreting out reasons for observed changes.
 A problem with the successive independent samples design occurs when the samples
drawn from the population are not comparable-that is, not equally representative of the
population. Although sophisticated statistical procedures exist to help unravel the
problems associated with noncomparable successive samples, the best solution is to avoid
the problem by carefully selecting successive samples that represent the same population
Longitudinal design: In the longitudinal design, the same respondents are surveyed over time in
order to examine changes in individual respondents. A longitudinal survey design involves the
survey procedure of collecting data about trends with the same population.
Example: Heatherton, Keel, and their colleagues have used the longitudinal design to investigate
changes in attitudes and behaviors related to eating during the transitions from college to early
adulthood and from early adulthood to middle years.
Human development from birth to old age.
Types:
 Trend Studies: In some surveys, researchers aim to study changes within some general
population over a period of time. This form of longitudinal research is called a trend
study. Trend studies are longitudinal survey designs that involve identifying a population
and examining changes within that population over time. A popular example of this
design is the Gallup Poll, which is used during elections to monitor trends in the
population of voters from the primary to the final election.
 Cohort Studies: A cohort study is a longitudinal survey design in which a researcher
identifies a subpopulation based on some specific characteristic and then studies that sub
population over time. For example, a cohort group of 18-year-olds is studied in the year
 2001. Five years later (in 2006), a group of 23-year-olds is studied. (They may or may
not be the same individuals studied in 2001.) Five years after that (in 2011), a group of
28-year-olds is studied. While the individuals studied each time might be different, they
must have been 18 years old in the year 2001 to qualify as representatives of the cohort
group.
 Panel Studies: A panel study is a longitudinal survey design in which the researcher
examines the same people over time. The high school seniors studied in 1998 will be the
same people studied in 2000, 1 year after graduation, and again in 2002, 2 years after
graduation. One disadvantage of a panel design is that individuals may be difficult to
locate, especially 2 years after graduating from high school. The advantage to this type of
study, however, is that the individuals studied will be the same each time, allowing the
researcher to determine actual changes in specific individuals.
Advantages: The longitudinal design has two important advantages.
 First, the investigator can determine the direction and extent of change for individual
respondents. Also, because changes in each individual’s responses are assessed, it’s
easier to investigate reasons for attitude or behavior changes.
 Second, the longitudinal design is the best survey design when a researcher wishes to
assess the effect of some naturally occurring event
Limitations:
 One possible problem is that respondents may strive courageously to be consistent across
surveys. This can be particularly troublesome if the study is designed to assess changes in
respondents’ attitudes! Although their attitudes have actually changed, people may report
their original attitudes in an effort to appear consistent.
 Another potential problem is that the initial survey may sensitize respondents to the issue
under investigation. For example, consider a longitudinal design used to assess students’
concern about crime on campus. Once the study starts, participants may pay more
attention to crime reports than they normally would. You might recognize this as an
illustration of reactive measurement—people behaving differently because they know
they are participating in a study.
Questionnaires and Interviews: A questionnaire is a form used in a survey design that
participants in a study complete and return to the researcher. The participant chooses answers to
questions and supplies basic personal or demographic information. An interview survey,
however, is a form on which the researcher records answers supplied by the participant in the
study. The researcher asks a question from an interview guide, listens for answers or observes
behavior, and records responses on the survey.
Mailed Questionnaires: A mailed questionnaire is a form of data collection in survey research
in which the investigator mails a questionnaire to members of the sample. A mailed
questionnaire is economical because it involves only duplication and mailing expenses.
Web-Based Surveys or Questionnaires: A Web-based questionnaire is a survey instrument for
collecting data that is available on the computer. Several software programs are available for
designing, gathering, and analyzing survey data with sample questions and forms.
One-on-One Interviews: In one-on-one interviewing in survey research, investigators conduct
an interview with an individual in the sample and record responses to closed-ended questions. In
one-on-one interviewing in survey research, investigators conduct an interview with an
individual in the sample and record responses to closed-ended questions.
Focus Group Interviews: In quantitative focus group interviews in survey research, the
researcher locates or develops a survey instrument, convenes a small group of people (typically a
group of 4 to 6) who can answer the questions, and records their comments on the instrument.
For example, this group might consist of parents who evaluate a new math or science curriculum
in a school.
In telephone interview surveys, the researcher records the participants' comments to questions on
instruments over the telephone. The researcher develops or locates an instrument, obtains the
telephone numbers of participants in the sample, conducts the telephone calls, and asks the
participants to answer questions on the instrument.
Instrument Design: Designing good survey instruments is a challenging and complex process.
When survey researchers design an instrument for data collection, they typically perform the
following steps:
1. They write different types of questions. These include personal, attitudinal, and behavioral
questions; sensitive questions; and closed- and open-ended questions.
2. They use strategies for good question construction. This includes using clear language, making
sure the answer options do not overlap, and posing questions that are applicable to all
participants.
3. They perform a pilot test of the questions. This consists of administering the instrument to a
small number of individuals and making changes based on their feedback.
Personal, Attitudinal, and Behavioral Questions
Background questions or demographic questions assess the personal characteristics of
individuals in your sample. These questions can be easy (i.e., gender) or difficult to answer (i.e..
level of income). Here are some examples of background questions:
What is your age?
How many years of teaching have you completed? (end of school year)
A second group of questions relates to obtaining individual attitudes or opinions from individuals
in your sample. For example, you might ask:
How much do you agree or disagree with this statement?
Most days I am enthusiastic about being a student.
Strongly agree
Agree
Neither agree or disagree
Disagree
Strongly disagree
A third group of questions can solicit information about the actual behavior of individuals in the
sample. For example:
Did you take a semester off during any of your 4 years of college?
Yes

No
Sensitive Questions; Some surveys contain sensitive questions that must be developed and used
with care.
Sensitive questions might have to do with:
Drug and alcohol use (e.g., use of cocaine)
Mental health issues (e.g., paranoid behavior)
Open-and Closed-Ended Questions:
In closed-ended questions in surveys, the researcher poses a question and provides preset
response options for the participant. A closed-ended question might be:
There are many reasons why adults wish to get more education. What is your most important
reason for coming to adult basic education classes? (Check one.)
To be able to help my children with their schoolwork
To get a better job
To improve myself
To get a high school equivalency diploma
Open-ended questions in a survey are questions for which researchers do not provide the
response options; the participants provide their own responses to questions. In an open-ended.
question, the participant supplies an answer.
For example:
Why are you attending adult education classes?
semi-closed-ended questions in a survey. This type of question has all the advantages of open-
and closed-ended questions. The technique is to ask a closed-ended question and then ask for
additional responses in an open-ended question. For example:
There are many reasons why adults wish to further their education. What is your most important
reason for coming to adult basic education classes? (Check one.)
To be able to help my children with their schoolwork
To get a better job
To improve myself
To get a high school equivalency diploma
Other (please comment)
Question Construction: Using good questions helps participants feel that they understand the
question and can provide meaningful answers. Good questions are clear and unambiguous, and
they do not confuse the participants. They also show respect for the participant by being sensitive
to gender, class, and cultural needs of participants. When you construct questions for a survey
questionnaire or interview, fit the questions to answers, include suitable response options, and do
not overlap. A review of these potential question construction problems and some solutions will
provide guidance for survey development.
The question is unclear. This usually occurs because words are vague or imprecise. Identify the
unclear or vague words and replace them with words understood by participants in the study.
There are multiple questions. Here, the question actually contains two or more questions, called a
double- or triple-barreled question. Reduce the multiple questions to a single question.
The question is wordy. When the question is too long, cut out unnecessary words to simplify and
shorten the question. Look for excessive use of prepositions (e.g., more than three) or qualifying
statements that lengthen the question.
◆ The question is negatively worded or wordy. If the question contains one or more negatives,
such as "should not," the meaning becomes unclear. Also, reword the question if it leads the
participants to one particular stance or another (e.g., using the word "pro-life"). Restate or
reword the question to eliminate negative connotations or leading words.
The question includes jargon. Jargon may not be familiar to all participants in a study. Eliminate
the jargon and use words familiar to all participants.
There are overlapping responses. This may lead to confusion when answering a question. Make
sure that the response options do not overlap by creating distinct options.
There are unbalanced response options. In this case, the responses may be unbalanced in
(e.g., "very important") and end with an "extent" word (e.g., "to a little extent"), rather than a
matching adjective (e.g., "not important"). Decide on a single response option and use it
consistently for all response categories for a question.
There is a mismatch between the question and the answers. The responses may not match the
"action" word used in the question. Identify the verb or adjective in the question that will be the
basis for the response options and create options using this word. (E.g., if the question says "to
what extent," the answer will say "a great extent.")
The question includes overly technical language. When this occurs, the respondent may not have
the level of understanding needed to respond to the question. Simplify the question so that all
individuals will know the meaning of the words and can respond to the question