Professional Documents
Culture Documents
Em (601) Report# 8
Em (601) Report# 8
SEMINAR
REPORT # 8
MOHAMED HAMMUDA
2120012
1 SAMPLING
Probability sampling
Non-probability sampling
2 Data Collection Methods
ASKING QUESTIONS
ACCOUNTS AND DIARIES
CONDUCTING INTERVIEWS
OBSERVING WITHOUT GETTING INVOLVED
IMMERSING ONESELF IN A SITUATION
DOING EXPERIMENTS
INTERNAL AND EXTERNAL VALIDITY
LABORATORY AND FIELD EXPERIMENTS
True experimental designs
Quasi-experimental designs
Pre-experimental designs
Ex post facto
3 MANIPULATING MODELS OR SIMULATIONS
diagramatic models
physical models
mathematical (or simulation) models
We needs a plan of action that identifies what data you need, where the necessary data are
to be found and what are the most effective and appropriate methods of collecting that
data. You will need to consider whether to get information from people, in single or large
numbers, or whether to observe and/or measure things or phenomena. You may need to do
several of these.
There are several basic methods used to collect primary data; here are the main ones:
• asking questions
• conducting interviews
• observing without getting involved
• immersing oneself in a situation
• doing experiments
• manipulating models
❖ Sampling
The mission to get all the information about all the subjects of research let it be
students or objects is nearly impossible and it requires a large budget to do so , so
the solution to this is taking a small group that represent the entire population of
data, and that is what we call (Sampling).
Case Study : A detailed study about a certain dynamic inside the sampling
group.
Sampling Frame : The certain groups that will be of interest of the study
within a population.
Difficulties are encountered if the characteristics of the population are not known, or if it is
not possible to reach sectors of it.
Characteristics of population :
The preference for a large sample must be balanced against the practicalities of the
research resources, i.e. cost, time and effort.
Sampling error : The measured difference between the mean value of a sample and that
of the population.
Bias is the unwanted distortion of the results of a survey due to parts of the population being
more strongly represented than others.
Probability Sampling:
This is based on using random methods to select the sample. Populations are not
always quite as uniform or one-dimensional as, say, a particular type of component in
a production run, so simple random selection methods are not always appropriate. We
should aim to guarantee that each element has an equal opportunity to be represented.
NON-PROBABILITY SAMPLING :
Types of questions:
• Asking people to relate their account of a situation or getting them to keep diaries is
perhaps the most open form of a questionnaire. These qualitative data collection
methods are used to find information on people’s actions and feelings by asking them
to give their own interpretation, or account, of what they experience.
• We must avoid leading questions, excessive guidance and other factors which may
cause distortion.
CONDUCTING INTERVIEWS
Interviews are more suitable for questions that require probing to obtain adequate
information.
• Structured interview
• Unstructured interview
• Semi-structured interview
OBSERVING WITHOUT GETTING INVOLVED
• This is a method of gathering data through observation rather than asking questions.
• This is a process of gathering primary data that not only involves observation, but also
experience in every sense of the word.
• Much of this kind of research is based on grounded theory which takes the approach of
collecting data in order to evolve theory rather than to test or refine an existing one.
DOING EXPERIMENTS
Data can be collected about processes by devising experiments, An experiment aims to isolate a
particular event so that it can be investigated without disturbance from its surroundings. They
are primarily aimed at gaining data about causes and effects – to find out what happens if you
make a change, why and when it happens and how.
• In order to be able to generalize the results beyond the confines of the experiment
itself, the experiment should really reflect the situation in the real world , it should
have :
• internal validity : the extent to which the ideas about cause and effect are supported by
the study.
Types of experiments
1. True experimental designs
2. Quasi-experimental designs
3. Pre-experimental designs
4. Ex post facto
A model, like an experiment, aims to isolate and simplify an event in order to inspect
it in detail and gain useful data. The difference is that models only provide a
representation of the event – a simulation – that shows relationships between
variables.
Types of models:
• diagrammatic models
• physical models
• mathematical (or simulation) models.