Professional Documents
Culture Documents
Lesson-2 Stats
Lesson-2 Stats
Data
DATA
MANAGEMENT Qualitative Quantitative
1 2
3 4
04/02/2023
Non-responsive sample subjects are a non-issue when More complex observations that ask observers to
you’re simply making direct observation. interpret something (e.g. how many cars are
driving dangerously) require more complex
training and are prone to bias.
5 6
• Questionnaires, as we consider
them here, are stand-alone
ADVANTAGES (Questionnaire)
instruments of data collection that
will be administered to the sample Questionnaires give the researchers an opportunity
subjects either through mail, phone to carefully structure and formulate the data
or online. They have long been one collection plan with precision.
of the most popular data collection
Respondents can take these questionnaires at a
techniques. convenient time and think about the answers at
their own pace.
7 8
04/02/2023
DISADVANTAGES (Questionnaire)
• Conducting interviews can
help you overcome most of
the shortfalls of the previous
Questionnaires without human intervention (as we have taken
them here) can be quite passive and miss out on some of the two data collection
finer nuances, leaving the responses open to interpretation. techniques that we have
Interviews and focus group discussions, as we shall see later, are
instrumental in overcoming this shortfall of questionnaires.
discussed here by allowing you
to build a deeper
understanding of the thinking
behind the respondents’
Response rates can be quite low
answers.
9 10
Interviews help the researchers uncover rich, deep Reaching out to all respondents to
insight and learn information that they may have conduct interviews is a massive,
missed otherwise.
time-consuming exercise that leads
to a major increase in the cost of
ADVANTAGES The presence of an interviewer can give the DISADVANTAGES conducting a survey.
respondents additional comfort while answering
(Interviews) the questionnaire and ensure correct
(Interviews)
interpretation of the questions. To ensure the effectiveness of the
whole exercise, the interviewers
The physical presence of a persistent, well-trained must be well-trained in the
interviewer can significantly improve the response necessary soft skills and the relevant
rate. subject matter.
11 12
04/02/2023
13 14
DISADVANTAGES
(FGD) The presence of excessively loud members in the focus group
• Usingdocuments and records canbeefficient and
inexpensivebecauseyou’repredominantly using
can subdue the opinions of those who are less vocal.
research thathasalreadybeen completed. However,
sincethe researcher haslesscontrolover the results,
documentsand recordscanbe incompletedatasource.
The members of a focus group can often fall prey to group-think
if one of them turns out to be remarkably persuasive and
influential. This will bury the diversity of opinion that may have
otherwise emerged. The moderator of a focus group discussion
must be on guard to prevent this from happening.
15 16
04/02/2023
Oral histories
Data
• Atfirst glance, an oral history might sound likean interview.
Bothdata collection methodsinvolveasking questions.But an
oral history ismore preciselydefined asthe recording,
preservation,and interpretation of historical information
based on the opinions and personal experiencesof people
whowere involved in the events. Qualitative Quantitative
17 18
19 20
04/02/2023
21 22
Sampling
• isthe processof identifying asubsetof apopulation that provides an accurate
reflection on the whole. Itcanbe a trickyprocess, aspopulations are often diverse.
However, there are some statisticalmethods that canmakesure a small subsetof the
community accuratelyrepresents the wholegroup.
Sampling Techniques
23 24
04/02/2023
• Justasitsname indicates,random sampling involves pickingrespondents withlikepicking • Easierthan random sampling, systematic sampling follows a
names outof ahat. While randomness mayseemunscientific,thismethod canbe setof rules to createregularity in sampling.Anexampleis interviewing everytenth
valuable in research, and in fact,isthe preferred wayof sampling, asatruly random customer. Aslong asyou follow the
counting system, you’ll knowthat there’ssome order to the
sample eliminates elements that canaffectthe validity of a study. process.
• Systematicsampling retains some of the benefits of
randomness, but it canbe too rigid in caseswhere, for example,the researcherknows
that the counting system is
going to result in data that skewsone wayor another — for
instance, if someone isinterviewing everytenth customer and everytenth person winds
up being afemale in the same agerange.
25 26
Convenience sampling
Purposive Sampling
• Thisisthe easiestsampling method but alsothe least reliable. Conveniencesampling • The purposive sampling technique, also called judgment
involves gathering information from whoever isclosestand easiestto reach.An example sampling, is the deliberate choice of a participant due to the
would be askingcoworkersin the sameoffice a question, rather than questioning every qualities the participant possesses.
employee atthe company,sincegoing to other officesmight takemore time and effort.
Conveniencesampling canalsoinvolve using whateverdata isreadily available.
27 28
04/02/2023
• With clusteredsampling, aresearcherusesthe subgroups of apopulation instead of • Another method that usessubgroups isstratifiedsampling. Thisdata collection
individuals. Clustersare often predefined, suchasmunicipalities in astudy about the method involves dividing apopulation into subgroups that share similar
effect of aparticular phenomenon acrossthe country. characteristics.
• Clustered sampling isfurther broken down into different types— single-stagecluster • For example, astudy canbreak respondents down by gender or age.When the
sampling, whereall individuals in aclusterare included in the sample, or two-stage components are easyto determine, like gender or age,the risk of biasislow,especiallyif
cluster sampling, whereonly random individuals within the cluster arechosen. the data comesfrom the respondents. Stratified sampling reduces bias, but sometimes
characteristicsare difficult to ascertain, whichcaneither frustrate the sampling processor
invite bias.
29 30
Methods of DataPresentation
31 32
04/02/2023
Methods of DataPresentation
Tabular Form
33 34
Methods of DataPresentation
Graphical
35 36
04/02/2023
37 38