Engineering Data Analysis

Topic: OBTAINING DATA

Data are facts and statistics collected together for reference or analysis.

Data collection is the process of gathering and measuring information on targeted

variables in an established system, which then enables one to answer relevant questions
and evaluate outcomes.

Qualitative and Quantitative Data

Qualitative data are data which cannot be ordered in a logical or natural way. Examples
given are the color of the eye, brand of detergent and names of president. There is no
logical reason to list blue eyes before brown eyes and etc.

Quantitative data represent measurable quantities. The values which these data can be
ordered in a logical and natural way. Examples of quantitative data are shoe sizes, height
of a person, number of students.

Primary and Secondary Data

Primary data is data that is collected by a researcher from first-hand sources, using
methods like surveys, interviews, or experiments. It is collected with the research project
in mind, directly from primary sources.

Secondary data is data gathered from studies, surveys, or experiments that have been
run by other people or for other research.

Methods of Data Collection

1. Surveys/Questionnaires
This is the most common method used in data collection. The data is collected by
asking questions or providing questionnaires to study participants. It can be self-
administered, administered by someone else and can be face-to-face, telephone, mail, or
web-based.

2. Pre/Post-Test
Surveys or measures are collected prior to an intervention among a target
population and then an intervention is implemented for a period of time before recollecting
the same survey or measurement data after the intervention is complete. The before and
after data is compared to detect changes that may be attributed to the intervention.

3. Existing Databases
This kind of secondary data is often used in conjunction with survey data. It
includes census data, knowledge/attitude/behavior (KAB) studies, criminal justice
statistics, performance data, non-confidential client information, agency progress reports,
etc.

4. Experiment
Experimental data is obtained in “controlled” settings. This can mean many things,
but essentially it is data which is generated by the researcher with full control over one or
many variables of interest.

Prepared by: swml

 Engineering Data Analysis

5. Observations
Observational data is data which is collected routinely, without a researcher
designing a survey or conducting an experiment. The researcher concentrates on what
is happening than directly asking the participant. This method is used to understand
behavior and the meanings attached to them.

6. In-depth Interviews
In-depth interviewing is a qualitative research technique that involves conducting
intensive individual interviews with a small number of respondents to explore their
perspectives on a particular idea, program, or situation.

7. Focus Group
A focus group is a small, but demographically diverse group of people and whose
reactions are studied especially in market research or political analysis in guided or open
discussions about a new product or something else to determine the reactions that can
be expected from a larger population. The use of focus groups is a research method that
is intended to collect data, through interactive and directed discussions by a researcher.

PLANNING AND CONDUCTING SURVEY

Before conducting a survey, the following steps should be completed.

1. Develop a clear and concise purpose statement – What the requester or researcher
wants to know and why they want to know it?

2. Develop the items (questions, etc.) for the instrument – Make sure they are clear,
unambiguous and bias-free.

3. Test the questions on a group of at least 20 volunteers to determine the face validity of
the items is adequate. Make any required changes in the items.
4. Develop the introduction to the survey and the letters that will precede the distribution
of the survey, accompany the distribution of the survey, and the reminders after the survey
has been sent out.

5. Determine the modality of the survey distribution (web, email, paper, etc.)

6. Schedule the sending of the initial letter, the launching of the survey instrument, and
the sending of the follow-up letter(s).

7. Avoid “coverage error” by gathering a sample list of potential participants that matches
the population of interest as closely as possible.

The following guidelines are provided for the preparation of survey items and all
correspondence surrounding the survey administration.

1. Where possible, have the most legitimate authority figure available sponsor the survey
and sign all of the correspondence.

2. Use a professional tone, a scholarly approach, and a business format for all survey
correspondence.

3. Write so as to demonstrate respect for participants’ time and regard for their opinions.

4. Set forth in very clear and concise language the purpose of the research and use of
the information being collected.

Prepared by: swml

 Engineering Data Analysis

5. Assure participants of their Confidentiality or Anonymity (whichever applies).

6. Be sure to clearly delineate how their voluntary participation in this survey can benefit
others (the entity, the profession, or society).

7. Stress the unique and limited opportunity that the survey provides them to affect
decisions, policies, procedures, etc.

8. Communicate the timeframes for data collection, analysis and reporting.

The steps below should be taken immediately after the survey administration phase of
the research has been completed.

1. Each survey received will be examined for completeness.

2. The entire database will be examined for completeness.

3. Decisions will be made on how to handle incomplete surveys and database fields.

PLANNING AND CONDUCTING EXPERIMENTS: INTRODUCTION TO DESIGN OF

EXPERIMENTS

1. State your question.

Example: Does fertilizer really help soil to grow more beans?

2. Formulate a hypothesis and make a prediction.

Your hypothesis might be: Yes, fertilizer helps soil grow more beans. Based on
your hypothesis your prediction might be: I predict the soil with the highest concentration
of fertilizer will result in more bean seeds germinating. Make sure your prediction can be
tested.

3. Choose the variables.

An independent variable is the factor that you will change in your experiment. For
the bean seed experiment the independent variable is the amount of fertilizer you put in
the soil of each test pot. The dependent variable is the factor that you predict will change.
For the bean seed experiment, it would be the number of seeds that grow in each test
pot.

4. Choose a control.
In a scientific experiment; the control is used as a standard of comparison. The
control is treated the same way as the other experimental group, except for the
independent variable. In the bean seed experiment, the control sample would not have
any fertilizer added to the soil.

5. Conducting the experiment and collecting data.

You must conduct your experiment in a careful manner to ensure that your data is
not flawed. In the bean experiment you would need to provide the same conditions for all
your test pots. They all would need to receive the same amount of sunlight and the same
amount of water. You would need to collect data at regular intervals. If you collect data
every day, you would need do it at the same time each day (for example, at 3:00 PM) for
each of the test pots. It is also a good idea to learn beforehand how to care for the plants.
Too much or too little water might kill them and your data would be flawed. Record the
details of your experiment and each day’s results in a chart, table or list. It is important to
record your data neatly and accurately. Take photos to document your investigation.

Prepared by: swml

 Engineering Data Analysis

6. Analyze your results Chart or graph the results of your experiment so that they can be
easily analyzed. For the bean seed experiment, you could make a bar graph. On the x-
axis (the horizontal axis), you would put your independent variable—the concentration of
fertilizer in each of your experimental samples. These are the numbers that you know in
advance. On the y-axis (the vertical axis), you would put your dependent variable. This is
the factor you will be measuring in your experiment—the number or height of bean seeds
that grow.

References:
https://en.wikipedia.org/wiki/Data_collection
https://www.ndcompass.org
https://betterthesis.dk/research-methods/lesson-1different-approaches-to-
research/collecting-data
https://www.statisticshowto.com/primary-data-secondary/
www.amnh.org
Introduction to Statistics and Data Analysis. Heumann, Christian and Shalabh, Michael
Schomaker

Prepared by: swml