x

Module 2
Collecting Data x
Using Numbers to x Lesson 1
Describe Data
Lesson 2
Data collection brings together raw
data, but this has to be processed to
give useful information. The amount
of detail in data sometimes obscures
underlying patterns and data
reduction clears away the detail and
highlights important features and
patterns. It enables a view of the
data that is concise, but still
accurate. There are two approaches
to summarizing and presenting data.
• To appreciate the importance of data collection
• To discuss the amount of data to be collected
• To discuss the aims of data reduction and presentation
• To design tables of numerical data
• To draw frequency distributions of data
• To use graphs to show the relationship between two variables
• To draw charts
• To appreciate the need for numerical measures of data
• To understand measures of location
• To understand the arithmetic mean, median and mode of
data
• To understand the purpose of index numbers
Collecting Data x

Framework is a formal structure of giving you a list of the

key elements that key factors we use in enterprise
architecture.

Data and Information

Managers need relevant information for their decisions. To
get this, they start with data collection, then process the data
to give information, and present the results in the best
formats. Data collection is essential in every organization,
because it starts the process of decision-making – and
without proper data collection, managers cannot make
informed decisions 1
Data collection does not happen by
chance, but needs careful planning.

1. Define the purpose of the data. 7. Design a questionnaire or x

2. Describe the data you need to other method of data collection.
achieve this purpose. 8. Run a pilot study and check
3. Check available secondary for problems.
data to see how useful it is. 9. Train interviewers, observers or
4. Define the population and experimenters.
sampling frame to give primary 10. Do the main data collection.
data. 11. Do follow-up, such as
5. Choose the best sampling contacting non-respondents.
method and sample size. 12. Analyze and present the
6. Identify an appropriate sample. results.
 Set the purpose of
the data and how it Planning Data
will be used
x
Collection

Decide what data is

needed to achieve
this purpose

Includes:
Design the data • the amount of data
collection • sources of data
• means of collection

Actually collect the

data
Amount of Data
Three important questions for data collection are the amount of data, its
source, and the means of collection.

A marginal cost as the extra In principle, you collect the

cost of collecting one more amount where the marginal cost
bit of data, and this rises with equals the marginal benefit.
the amount of data collected. If you collect less than this, you
lose potential benefit because the
The marginal benefit of data cost of collection is less than the
– which is the benefit from the benefit; if you collect more data
last bit collected – falls with than this, you waste resources
the amount collected. because the cost of collection is
more than the benefit.
 Marginal
cost of data
value

Finding the Amount x

of Data to Collect

too much
data, as the
too little cost is
data, as the greater than
benefit is benefit
greater than
cost

Marginal
benefit of
the data

Optimal amount Amount of data

of data collected
Types of Data
Nominal Data
The kind that we really cannot quantify with any meaningful
units.

Ordinal Data
One step more quantitative than nominal data. Here we can
rank the categories of observations into some meaningful
order.

Cardinal Data
Has some attribute that can be measured directly.
Primary and Secondary Data

Primary Data
This is often characterized as field research when you actually
go out and collect data yourself. It is a new data collected by
an organization itself for a specific purpose.

Secondary Data
Characterized as desk research when you look for data that
someone else has already collected. It is an existing data that
was collected by other organizations or for other purposes.
Using Samples to Collect Data

A population consists of all the people or items that

could supply data. These are listed in a sampling
frame. It can be difficult to choose the right
population and find a sampling frame.
It is usually too expensive, time-consuming and
difficult to collect data from the whole population –
giving a census. The alternative collects data from a
representative sample of the population and uses
this to estimate values for the whole population.
 SIMPLE SYSTEMATIC
RANDOM SAMPLE
SAMPLE

STRATIFIE
D CLUSTER
SAMPLE SAMPLE
Types of x
Samples
 Quota x
Sampling

Quota:
Male above 50
Organizing Data Collection

Observation
A way of gathering data by watching behavior, events,
or noting physical characteristics in their natural
setting. It can be overt or covert. It can also be either
direct or indirect.

Questionnaires
Can be administered through personal interview,
telephone interview, the Internet, postal survey, panel
survey or longitudinal survey.
How to Plan for x
Observations

• Determine the focus. Think about the evaluation question(s)

you want to answer through observation and select a few areas
of focus for your data collection.
• Design a system for data collection. Think about the specific
items for which you want to collect data and then determine
how you will collect the information you need. There are three
primary ways of collecting observation data: recording sheets
and checklist; observation guides, and field notes.
• Select the sites. Select an adequate number of sites to help
ensure they are representative of the larger population and will
provide an understanding of the situation you are observing
How to Plan for x
Observations

• Select the observers. You may choose to be the only observer

or you may want to include others in conducting observations.
• Train the observers. It is critical that the observers are well
trained in your data collection process to ensure high quality
and consistent data.
• Time your observations appropriately. Programs and
processes typically follow a sequence of events. It is critical
that you schedule your observations so you are observing the
components of the activity that will answer your evaluation
questions. This requires advance planning.
 Guidelines for Designing x
Questionnaires

• A questionnaire should ask a series of related questions and

should follow a logical sequence.
• Make the questionnaire as short as possible.
• Questions should be short, simple, unambiguous, and easy to
understand and phrased in everyday terms.
• Even simple changes to phrasing can give very different results.
• Avoid leading questions. Such questions encourage conformity
rather than truthful answers.
• Use phrases that are as neutral as possible.
• Phrase all personal questions carefully.
 Guidelines for Designing x
Questionnaires

• Do not give warnings – a question that starts ‘We

understand if you do not want to answer this, but . . .’ will
discourage everyone from answering.
• Avoid vague questions.
• Ask positive questions rather than a less definite one.
• Avoid hypothetical questions. Any answer will be
speculative and probably not based on any real thought.
• Avoid asking two or more questions in one.
Guidelines for Designing x
Questionnaires

• Open questions collect general views, but they favor the

articulate and quick thinking, and are difficult to analyze.
• Ask questions with preceded answers, with respondents
choosing the most appropriate answer from a set of alternatives.
There are many formats for these.
• Be prepared for unexpected effects, such as sensitivity to the
color and format of the questionnaire, or different types of
interviewer getting different responses.
• Always run a pilot survey before starting the whole survey. This is
the only way to identify problems and improve the questionnaire
design
 Guidelines for Designing x
Questionnaires

• failure to identify an appropriate population

• choosing a sample that does not represent
this population
• mistakes in contacting members of the
sample
• mistakes in collecting data from the sample
• introducing bias from non-respondents
• mistakes made during data analysis
• drawing invalid conclusions from the analysis
Using Numbers To Describe Data x

MEASURING DATA
Numerical measures of data give more objective and accurate
descriptions.

Measure of Location
to show where the center of the data is, giving some kind of
typical or average value
Measure of Spread
to show how the data is scattered around this center, giving
an idea of the range of values.

2
MEASURE OF LOCATION

Arithmetic Mean x
• the most widely used
measure, giving an
average value
• add all the values
together to get the sum
• divide this sum by the
number of values to get
the mean.
MEASURE OF LOCATION Mode x
• drawing a frequency
distribution of the
data
Median x • identifying the most
• arranging the values frequent value –
in order of size which is the mode.
• counting the number
of values
• identifying the
middle value – Choice of Measure x
which is the median. • the mean is the simple average
• the median is the middle value
• the mode is the most frequent
value.
Advantages and Disadvantages

MEDIAN x
Advantage
• Simple to understand and easy to calculate
• The middle part of the set, not affected by the extreme
values.
Disadvantage
• Need to be arranged, if the set contains a large
amount of quantity, the process is slow.
• Cannot be rigidly defined, if you happen to have even
number of items, median cannot be exactly found.
Advantages and Disadvantages

MODE x
Advantage
• Simple to understand and easy to calculate
• Not affected by extremely large or small values.
• Can be located just by inspection
• Useful for qualitative data
Disadvantage
• Not well defined
• Not based on all the values
• Not capable of further mathematical treatment
• Sometimes the data has one or more than one mode
and sometimes the data has no mode at all.
 MEASURE OF SPREAD

Range and Quartiles x

• The simplest measure of spread is the range, which is the
difference between the largest and smallest values in a set
of data.
• The broader the range, the more spread out the data.
 MEASURE OF SPREAD

Mean Absolute Deviation x

• The deviation is the difference between a value and the
mean.
• A basic measure gives the mean absolute deviation.
Alternatively, we can square the deviations and calculate
the mean squared deviation – or the variance.
 MEASURE OF SPREAD

Variance and Standard Deviation x

The square root of the variance is The standard deviation is used for
the standard deviation, which is other analyses, such as the
the most widely used measure of coefficient of variation – gives a
spread. We can usually estimate relative view of spread, and the
the number of observations coefficient of skewness – describes
expected within a certain number the shape of a distribution.
of standard deviations of the
mean.