1. The techniques used to study and provide a statistical data is called
observational techniques.
2. What is an Observation?
The act of making and recording a measurement is called observation.
Observational Research:
Observational research consists of systematic observation.
It draws inferences about the possible effect of a treatment on subjects where the
subjects of treated group and control group is outside the control of the
investigator.
Observational study consist of two components:
1. Retrospective:
Observation of the event that happened in past.
2. Concurrent:
Observation of the event that is happening currently.
In observational study we have:
1. Stimulus variable:
A variable that is being observed
Example: Teaching style
2. Response option:
The description of the trait:
Example: good, bad, excellent.
3. In studies of children, observational techniques are highly developed.
Experiments that might alter children's lives would raise ethical
problems.
Observational techniques can be done without disturbance.
Following are examples of different observational techniques,
Naturalistic observation: (that takes place in a natural or everyday setting
such as a school.)
Controlled observation :(observational research, carried out under
carefully arranged conditions. Each subject is exposed to the same
situation, to see differences between individual reactions.)
Standardized testing :( is a form of controlled observation using testing
procedures previously shown to be reliable and valid.)
Clinical observation:(observations made by a skilled clinician interacting
with a patient or client. The clinician takes notes on the interaction,
usually immediately after the interview or meeting with the client.)
4. Surveys and polling:
In survey and polling, data are collected from large numbers
of subjects. A survey can be about anything. A poll usually
asks for opinions or value judgments. The goal of both is to
determine the characteristics of a larger population from a
relatively small sample.
◦ For example, you might conduct survey on a random sample
of students from your campus to determine their
attitudes towards various activites.
Interviews:
◦ Interviews are structured conversations (that is, they
follow some pre-arranged plan or pattern). Interviews
can be combined with survey methods.
◦ For example, you might interview a random sample of
preschoolers from a town where a newsworthy event took
place, asking each child carefully worded questions to
determine his or her perception of the event.
5. Microanalysis:
◦ It is a detailed analysis of very brief events.
Sometimes researchers notice interesting things
simply by slowing down a quick movement. This
can be done with video cameras set to capture
many images per second.
For example:
◦ we can observe the body language more effectively if it is slow down with
the help of a video camera.
6. Rating Scales:
Rating scales is a technique in which the observer or Rater
categorize or rate the objects, events, or behavior of a person by
the series of continuous numerals.
In Rating scales we observe the actual behavior and the
remembered behavior.
◦ Behavior:
The aggregate of the responses or reactions or movements made by an
organism in any situation.
It may be classified into two components:
1. Actual behavior:
The behavior that is physically present.
2. Remembered behavior:
The behavior shown in past.
7. There are four types of Rating Scales:
1. Nominal Scale
2. Ordinal Scale
3. Interval Scale
4. Ratio Scale
Nominal Scale:
◦ It is the lowest level of measurement in which we place data into
categories, without any order or structure.
Characteristic:
It is the most simple scale.
It has no order or arrangement.
Nominal scale is used to identify the objects only.
In research activities a YES or NO scale is Nominal.
Example:
◦ How many students are there in AWKUM?
Suppose there are : 1000 students
When categorized: 500 male and 500 female. (Nominal Scale)
8. Ordinal scale:
The orderly arrangement of objects is called ordinal
scale. (Label+order)
The simplest ordinal scale is ranking.
◦ Characteristics:
An ordinal scale only lets you interpret gross order and
not the relative positional distance.
This arrangement does not mean that one object is better
than the other.
Example: which shampoo is commonly used in Mardan?
1. Clear 2. Head & shoulder 3. Susilk
◦ In this example we don’t mean that which product is best, we just rank its
usage. This is just an ordinal scale of preference.
9. Interval scale:
◦ In interval scale we categorize the data with its
intervals. (label+order+inerval)
Characteristic:
The interval may be zero, because in interval scale the zero is
called arbitrary zero, which means the zero also has some
value.
For instance; a child gets zero marks in a subject, so it doesn’t
mean that he has no knowledge, he has some knowledge, but
due to some reasons he could not get marks.
Example:
◦ we have some data,, 2,4,6,8, this data has an interval of 2.
If its 0,1,2,3,4,5,6,7,8,9 this data has also an interval of 1.
10. Ratio scale:
◦ The factor which clearly define Ratio scale is the
True zero. (Label+order+interval+true zero)
Characteristic:
◦ Ratio scale is the top level of measurement and is
not often available in social research.
◦ It is mostly used in experimental research.
◦ In ratio scale the important thing is the true zero.
Example:
◦ If the weight is zero .. It means true zero…Nothing
◦ If the length is zero .. It means true zero .. Nothing
11. Observational techniques avoid the self-selection bias that often
distorts the data gathered by techniques such as surveys.
Observational techniques In any group presented with a survey, the
people that choose to participate often fail to represent the group
adequately. This often captures both those that willingly participate
in non-observation research approaches and those who do not
normally participate
Observational techniques can also provide a depth of information
lacking in other techniques, through questions prompted by the
observation itself.
For example:
◦ A researcher observes a customer pick out a name-brand detergent and then opt for
a less expensive brand, the researcher can ask if the decision was financial or about
the intended use.
12. It employs technology such as cameras and recording devices,
observational research represents a larger financial investment than
any other research methods.
A true ethnographic study requires the participation of someone
with advanced anthropological training over a period lasting from
days to weeks.
Data analysis can be time consuming, and observational research
often fails to provide insight into attitudes and motives.
For example:
◦ A recording may show you how long a customer spends reading a package, but it
probably can’t provide you any insight into what, if any, portion of the packaging
contributed to a buying decision.
13. The qualitative nature of the data gathered by observational
research limits how much the information generalizes.
Everything from community norms to odd store design can
influence the results.
Observational results often prove more effective when paired
with data gathered from other research methods, such as
focus groups and surveys.
Multiple sources of data can help to level out skewed results,
and can even provide insight into the reasons.
Customers choose to represent their decisions in a particular
light.
The End.