Data Gathering Procedure – Observation Method

Sources of Data – Interview Method

• What is data? Data is a existing – Collection Of Data Through
information /knowledge represented or Questionnaires
coded in some form suitable for better
Sources of primary data:
usage or processing.
• Observation Method 2 Types of
• Data is a set of values of qualitative
Observation:
or quantitative variables.
– Participant Observation. The
• Quantitative data are anything that
principal investigator engaged in activities
can be expressed as a number, or quantified.
of the group watched. The members may or
These data may be represented by ordinal,
may not be aware that they are being
interval or ratio scales and lend themselves
observed.
to most statistical manipulation.
– Non-Participant Observation. The
• Qualitative data is a categorical
principal investigator does not engage in
measurement expressed not in terms of
activities of the group being observed. The
numbers, but rather by means of a natural
researcher just watches and listen in the
language description. In statistics, it is often
group’s activities and draw conclusions.
used interchangeably with "categorical"
data. Sources of primary data:
For example: favorite color = "blue" 2. Interview. This is defined as a person
to person communication (Kumar, 2011). It
can also be through telephone/mobile and
Sources of Data internet is also emerging as an interview tool
nowadays forms between two or more
Primary and Secondary Data
persons with a definite purpose in mind. It is
While deciding about the method of data used conversation used to gather information
collection to be used for the study, the whereby the researcher will manage the
researcher should keep in mind two types of process of conversation and ask questions
data - primary and secondary. while the respondents will answer the
questions. There are 2 types – unstructured
Primary data are those which are collected and structured.
afresh and for the first time and thus
happens to be original in character. It is the Sources of primary data:
real time data which are collected by the
2 Types of Interview
researcher himself. A “first hand data”
a. Unstructured Interview. The
Secondary Data – are those in which
researcher has the complete autonomy to use
information has already been published from
words, content, structure of the interview
different sources like books, online portals,
and sequence of questions according to the
journals, newspapers, magazines and other
context of the study. this is usually used in a
sources.
qualitative research.
b. Structured Interview. The researcher
Data Collection Methods asked a fixed set of questions during the
interview. This is good as it gives a
• The methods of collecting data consistent data which guarantee the
mainly refers to collecting primary data, comparability of data
because secondary data are already
available. Sources of primary data:

• Sources of primary data are:

3. Questionnaire. A written catalog of •The root word of the word variable is ‘vary’
questions projected to provide answers to or ‘can change’
the research problems. The content of the
questionnaire has to undergo content and
construct validation before it can be Definitions
administered to the respondents. The aim of
the questionnaire is to receive a comparable •Variable:“any entity that can take on a
answer from all the respondents. variety of different values” (Wrench et al,
2008, p. 104)
–gender
Exercise 1: Choose the letter of the correct
–self-esteem
answer.
–agreement
__1. What is “first hand” data?
•An attribute, on the other hand, is a specific
A. Primary data B. Secondary data value on a variable. Ex:
__2. An example of secondary data.
–Variable: gender, the attributes: male and
A. Field inquiry B. Journals female
__3. Data that is expresses by means of a –Variable: self-esteem, the attributes
natural language description is according to level: High, Medium, Low
A. Quantitative data B. Qualitative –Variable: agreement, the attributes
data according to level: strongly disagree,
disagree, neutral, agree, strongly agree
___4. This type of data collection method
requires a person to person communication.
A. Interview B. Questionnaire Varieties and types of variables
Nominal variables represent categories
that cannot be ordered in any particular way.
Designing a Questionnaire
–religions; Christians, Muslims, Jews, etc.
A questionnaire is an instrument for
collecting data. It consists of a series of –marital status; single, married, divorced
questions that respondents provide answers
to a research study. –sex/gender: males vs. females

Steps in Designing a Questionnaire: Interval variables have values that lie

along an evenly dispersed range of numbers
Step 1 – Background
–Temperature
Know your variable or construct.
–Blood pressure
Ordinal variables represent categories
Variables that can be ordered from greatest to
smallest.; with an order, sequence, or
• Definition: Variables are properties
hierarchy
or characteristics of people or things that
vary in quality or magnitude from person to –fall, winter, summer, spring
person or object to object (Miller &
Nicholson, 1976). –K-6, junior high, high school, college

•The most common variables in social Ratio variables possesses the properties
research are age, sex/gender, education, of interval variable and has a clear definition
income, marital status and occupation. of zero, indication that there is none of that
variable.
-height, weight, and distance. (independent); sea level (dependent) is
influenced by temperature (independent)
Exercise 2: Write I if the variable is Interval,
N if Nominal, R if Ratio, and O if Ordinal. Example:
___1. Military title •Effectiveness of aroma therapy to the
reading comprehension of high school
___2. Temperature in degree Farenheit
students.
___3. Birthplace
•Dependent variable: reading
___4. Year level comprehension

___5. Type of music •Independent variable: aroma therapy

___6. Clothing item

___7. A score in 10-item quiz Exercise 3: Write IV for independent
variable and DV for dependent variable
___8. Feeling for today
•Phenomenon A: Climate Change
___9. Means of transportation
1.Sea level ___
___10. BP of 110/70 mmHg
2.The amount of rainfall ____
•Phenomenon B: Crime & Violence on
Independent variable streets
•the variable that is manipulated either by 3. Number of law enforcers ___
the researcher or by nature or circumstance
4. Number of robberies ____
•independent variables are also called
“stimulus” “input” or “predictor” variables •Phenomenon C: Performance of students in
college entrance exams
•analogous to the “cause” in a cause-effect
relationship 5. Number of hours devoted to studying ___

•Examples are age, gender, what people eat , 6. Entrance exam score ___
how much time they spend using gadgets,
how much television they watch
Designing the Questionnaire
Step 2 –Questionnaire Conceptualization
Dependent variable
•Choose the response scale to use. How the
•a variable that is observed or measured, and
respondents will answer the questions in
that is influenced or changed by the
your study.
independent variable
Likert Scale sample ratings:
•dependent variables also called “outcome
variable” is the result or effect of the
changes brought about by another variable
(usually independent variable).
•analogous to the “effect” in a cause-effect
relationship
•This is what the researchers are interested
in.
•Examples are: test scores (dependent)
depends on how much you studied
Designing the Questionnaire Face validity –this is superficial
assessment. The questionnaire appears to
Step 2 –Questionnaire Conceptualization
measure the construct or variable that the
•Generate the items or questions of the research study is supposed to measure.
questionnaire based on the purpose and
Content validity –experts are asked to
objectives of the research study. Keep in
provide feedback on how well each question
mind the ff:
measures the variable or construct under
–Questions should be clear, concise and study. The experts make judgments about
simple using minimum number of words the degree to which the test items or
statements match the test objectives or
–Classify your questions under each specifications
statement based on your problem statement
Criterion-related validity –measures the
–Questions should be consistent with the relationship between a measure and an
study outcome.
–Avoid sensitive or highly debatable Construct validity –concerned with the
question. extent to which a measure is related to other
Step 2 –Questionnaire Conceptualization measures as specified in a theory or previous
research.
•Choose the types of questions in developing
statements: Designing the Questionnaire

Dichotomous question –ex. Yes/No Step 3 –Establish the Validity of the

Questionnaire
Open-ended question –usually answers the
question “why” Validityis traditionally defined as “degree to
which a test measures what claims, or
Closed questions – purports, to be measuring” (Brown, 1996)
Multiple-choice questions, ex: What is the Ways to assess the validity:
highest level of education your mother has
completed? _ elementary _high school Face validity –this is superficial
_college _don’t know assessment. The questionnaire appears to
measure the construct or variable that the
Rank-order scale questions –respondents research study is supposed to measure.
are ask to rank their choices on each
statement or item. Ex: Pls. rank the ff. Content validity –experts are asked to
activities in order of importance in your provide feedback on how well each question
work as a senior high school student. (Plsuse measures the variable or construct under
“5” as the most important activity, “1” is the study. The experts make judgments about
least important activity. the degree to which the test items or
statements match the test objectives or
Rating scale question –you construct a specifications
scale like those examples given for
Likertscale ratings. Criterion-related validity –measures the
relationship between a measure and an
outcome.
Designing the Questionnaire Construct validity –concerned with the
Step 3 –Establish the Validity of the extent to which a measure is related to other
Questionnaire measures as specified in a theory or previous
research.
Validityis traditionally defined as “degree to
which a test measures what claims, or Designing the Questionnaire
purports, to be measuring” (Brown, 1996) Step 4 –Establish the Reliability of the
Ways to assess the validity: Questionnaire
Reliabilityindicates the accuracy or •Self-administered surveys
precision of the measuring instrument
Respondents complete on their own
(Norland, 1990). It refers to a condition
where measurement process yields Best designed for measuring variables
consistent responses over repeated with numerous values or response categories
measurements.
Investigating attitudes and opinions not
Ways to assess reliability of a questionnaire: usually observable
Test-retest reliability –the same Describing characteristics of a large
questionnaire is administered twice and population
correlation between the two sets of scores is
computed. Response rates tend to be lowest for
mailed questionnaires, as low as 20 –30%.
Split-half method –also called equivalent Low response rate affects generalizability
or parallel forms. In this method, two
different tests covering the same topics are More standardization of the questions
used and the correlation between the two Response patterns can vary
sets of scores is calculated.
Not suitable for young children, visually
Internal consistency (inter-item)– impaired, with learning/reading disabilities
reliability of questions are measured on an
interval or ratio scale. One popular formula Designing the Questionnaire
used to measure internal consistency is Forms of Questionnaire:
Cronbach’salpha, which ranges from 0 (poor
reliability) to 1 (perfect reliability). •Face-to-face interviews
Anything above .70 is considered Response rates tend to be highest with
sufficiently reliable. face-to-face interviews
Designing the Questionnaire Excluding ‘street corner’ technique
Step 5 –Pilot Testing the Questionnaire More time
Pilot testing is very important before you use Smaller samples
it to collect data. Through this process, you
can identify questions or statements which Higher cost
are not clear to the participants or there Two types of interviews: unstructured &
might be some problems with the relevance structured
of the questionnaire to the current study. 10-
15 people are needed for pilot testing. Designing the Questionnaire

Designing the Questionnaire Forms of Questionnaire:

Step 6 –Revise the Questionnaire • Telephone surveys

After identifying the problem areas in your Most popular

questionnaire, revise the instrument as
Less costly
needed based on the feedback provided
during the pre-testing or pilot-testing. The Less time
best questionnaire is one that is edited and
Less subjective to interviewer as compared
refined towards producing clear questions
to face-to-face
arranged logically and in sequential order.
The questionnaire should match with the Often conducted with computers
research objectives.
Can probe for information/clarification
Designing the Questionnaire
Threshold about 20 minutes
Forms of Questionnaire:
Designing the Questionnaire
Forms of Questionnaire:
•Computer assisted Surveys
Way to create and administer self-
administered questionnaires
Marketing researchers find response rates
increase
Questionnaires (especially short ones) can
be sent via email or provide internet link to
site which hosts survey
•Web-Based Surveys
Create own web page or hire a commercial
company (example:
www.hostedsurvey.com)
Allow for instant data coding
Need to be able to write code or use
software
Maybe less time and costs
Access is a huge issue (affecting
generalizability)
Internet access

Exercise 4: Designing a Questionnaire

•Construct a closed question related to your
career choice in college. (1 pt.)
•Construct an open-ended question related to
your career choice in college. (1 pt.)
•Formulate a dichotomous question related
to the following: (3 pts.)
–Gender
–Death penalty
–Anti-abortion law