Principles of survey research

LOGO

Principles of Survey ResearchPrinciples of Survey Research

Author:O. Tamimi

An Najah National University

Contents

What is survey?Example of surveysThe process of conducting a surveyData gathering techniques Types of data collectedTypes of survey questionsSurvey sampling procedure Data analysis & discussion

What is survey?

A survey is a research method for collecting information from a selected group of people using standardized questionnaires or interviews. Questionnaire is just one part of survey Process What is questionnairesA list of survey questions asked to respondents to extract specific information. Questionnaire is the data collection component of overall survey.

Are questionnaires and surveys the same ?Very often the questionnaire and survey are used to mean the same thing. But there is an important distinction between them: Survey describes the process of conducting the research, which includes a series of 7-steps, while the questionnaire is one part of the survey process

Page-1 An Najah National University O. Tamimi

Examples of Surveys


General Topics Specific Survey Studies

Course evaluation survey Junk food health related issues survey

Classrooms evaluation survey Health consequences of smoking cigarettes survey

Product Survey Hazardous chemicals in our everyday lives survey

Customer satisfaction survey Gender and racial discrimination at work place survey

Event satisfaction survey Al-Najah Campus safety and security survey

Health assessment survey Taxi driver survey in Nablus city

Patient satisfaction survey

The process of conducting a survey

Step-1: Defining the objective of the survey. It often comes from the thought “why I am doing this survey?”

Step-2: Constructing the questionnaire, a well designed questionnaire can provide information about opinions, beliefs & practices of group of individuals

Step-3: Methods of conducting the survey Decide about the procedure Sample size, how many people you will survey and their age How you will survey your subject(by phone, in class or interview) Step-4: Testing the questionnaire

Find out if the questionnaire’s content is clear and consistent with the subject. Get some of your colleagues to review them

Step-5: Administering the questionnaire in small study you can administer the questionnaire your self, but in large survey you will need helpers Step-6: Data entry and analysis statistical analysis is to be carried out Step-7: Writing the report The final stage of survey process is documenting the findings in a report



There are two main basic ways in which survey data are gathered: Interview based and Questionnaire based

Data Gathering Techniques (Survey Classifications)

Interview over the phone

Face to FaceInterview

Self Completed

Hand DeliveredQuestionnaire

Web Questionnaire

Interview Based

Classifications of Surveys

Hand Delivered Questionnaire (self completed questionnaire). The questionnaire is handed in person, the respondent will fill it up and submit it back to the researcher

Advantage Less expensive compared with the face to face interview

DisadvantageLimited opportunity for clarification if a respondent doesn’t understand a question

Web Questionnaire A collection of questions presented on a webpage and sent to the respondent’s email addresses. The survey design, distribution, completion and data analysis are done using through web-based applications

Advantage Easy to administer, convenient for respondents

Disadvantage Can be confused as SPAM

Face to Face Interview (questionnaire is completed by the interviewer) The interviewer reads the questions in a face to face setting

Advantage Interviewer is present to provide assistance

Disadvantage High cost because it involves a lot of traveling and hiring interviewers

Phone Interview Interviewers usually conducts a phone interview with the respondents. The survey questions and their answers are handled by the interviewer

Advantage Large number of surveys can be acquired in a short time

Disadvantage Requires a highly skilled interviewers


Face to FaceSurvey

TelephoneSurvey

Self CompletedQuestionnaire

WebQuestionnaire

Turn Around Time Slow Fast Fast Fast

Cost High Medium Low Very Low

Interviewers Required

Yes Yes No No

Response Rate High Medium Low Low

Characteristics of Data Gathering Techniques


cost: refers to the overhead cost and travelling cost response rate: refers to the number of people who answered the questionnaire divided by the number of people in the sample

Notes

There are four main types of data you might want to obtain through the survey

Types of Data Collected

Attitude To indicate whether the respondents have positive or negative feeling about an issue

Behavior

Asking the respondents to describe what they have done in the past or what they are currently doing. They are typically analyzed in terms of Whether the behavior is present or absent Frequency of the behavior

Opinion It is the respondent’s thought towards something

AttributesDemographic

Data

Typically ask for information about the characteristic of the population: like age, gender, ethnic-group, education, occupation, or cultural background. Attribute questions are important to evaluate how attitudes, behavior differ for respondents with various attributes

Fact: I learned that there was a small earthquake hit the city of NablusOpinion: A major earthquake is expected to take place

To what extent do you think that indirect exposure to smoke is dangerous for health?

What is your gender? How old are you? What is your race/ethnicity?

In a typical weak, how many times do you eat out at a restaurant? (eating habit)


Types of survey questions

Open ended question: is a question at which there is no single answer.it actually gives the respondents the opportunity to answer using their own words, the answer could come in one sentence or more.Closed ended question: has a definite set of answers from which the respondent choses. closed ended question is easy to summarize, but it may take much more time to design. Closed ended questions have many different styles, these are: Multiple choice question Categorical question Dichotomous question Rank order question Likert scale question (Matrix)


Open ended question Closed ended question

Advantage They permit adequate answers to complex questions

It is easier to answer and easier to analyze

Disadvantage A greater amount of respondents time, thought and effort is necessary

Take more time to construct

Example Why did you choose your major? Are you satisfied with your college?

Multiple Choice Question (single or multiple responses)

The respondent is given a list of items and is asked to choose either one answer that best applies or multiple answers Types of multiple choice questions Radio buttons or check-boxes Drop-down Matrix Questions

Categorical Question

Example (single response using check-boxes)What is your opinion of the product you recently purchased?

ExcellentGood

PoorFair

Used to categorize the respondents according to their age, gender, ethnic group or their level of education.

18-2819-28

39-4829-38

Which age range you are in:Example


Dichotomous Question Dichotomous questions are straight forward, typically asking respondents to choose between Yes / NoAgree or DisagreeTrue/False

Have you ever purchased a product or service from our website? Yes No

Example

Rank order question Rank order questions allow a certain set of brands or products to be ranked based upon a specific characteristic. Such as; cost, usage, reliability or features

ExampleBased upon what you have heard or experienced, please rank the following car brands according to their reliability.

2 Honda 4 Toyota 1 Mazda 3 Ford


Likert scale question (Matrix)

Strongly Disagree

Disagree Neutral

Agree Strongly Agree

I do not understand the concept behind the Lab experiment ●I like labs where I get a full help in doing the experiments ● It was clear how the Lab experiments fit into this course

●

The lab manual for this course was well written ●

Tick one box to indicate your agreement or disagreement with each statement

Example

It allows the respondents to rate a specific question on a scale of choices from i.e. strongly-agree to strongly-disagree. These choices are shown horizontally by default


Survey Sampling Procedures

General sampling techniquesRandom Sampling: every member of the population has an equal chance of being selected

Advantage: Simplicity Disadvantage: Not accurate if there is a variation

Example: in the following diagram, 13 units are randomly sampled from the survey population of 100 students.

Note: No clustering is needed for the population sample

The term Refers to the selection of individual, units or cases to be studiedIf the population of the people you are interested in is small enough, it may be possible to survey the entire population, rather than choosing a sample. But if you have a large population, you may have to select a sample

Factors that influence sample sizeThe size of sample for a particular survey depends on many factors Cost consideration (Budget, Desire to minimize the cost) Administrative concerns (Research deadline) Acceptable level of accuracy


Stratified Sampling: is an accurate sampling method which involves dividing the surveypopulation into specific subgroups (strata), the population of each strata shares the samecharacteristic such as; age, gender, career, income.

Reasons for stratified sampling To ensure that all parts of the population are highlighted in the sample

Example: in the following diagram, the population survey consists of 50 students, 25 of whom were females and the remaining are male.

Step-1 in stratified sampling the population are splitinto two strata. Female strata and Male strata. Step-2 random sampling is carried out twice; one for males,and one for females Step-3 since the two strata have an equal size, then the size of each sample would ideally be equal

Sampling size: if the population (N) consists of male-strata (m) & female-strata (f), where N= m+f. if the total number of sample is (S), male-sample size (s1) s1= m x (S÷N) female-sample size (s2) s2= f x (S÷N)

Advantage: Accuracy Disadvantage: Requires prior information about the population being sampled


Strata-1

Male, Full-timeSize:40

Strata-2

Male, Part-timeSize:20

Strata-3

Female, Part -timeSize:16

Strata-4

Female, Full-timeSize:24

Population Survey: N=100

Example: The survey population consists of 100 worker divided into 4-strata, each of which has a different size. If the required sampling size (S=25)Calculate the sample-size for each strata Sample-size (strata-1) s1= 40x(25÷100) =10Sample-size (strata-2) s2= 20x(25÷100) = 5Sample-size (strata-3) s3= 16x(25÷100) = 4Sample-size (strata-4) s4= 24x(25÷100) = 6

Notice: S=s1+s2+s3+s4


Data Analysis & Discussion

Once the questionnaires are completed and collected, the first step is to interpret the data on excel sheet to create specific drawings or charts.

A questionnaire usually comes with a collection of closed ended questions like: multiple choices and scaled matrix questions. For easier analysis each item in these questions is converted into numerical value For example: The scaled question (e.g. strongly agree, agree, neutral, disagree & strongly disagree) shall be converted to these values (1,2,3,4,5)

Strongly Disagree

Disagree Neutral

Agree Strongly Agree

I do not understand the concept behind the Lab experiment ●I like labs where I get a full help in doing the experiments ● It was clear how the Lab experiments fit into this course

●

The lab manual for this course was well written ●


Questions Strongly Agree Agree Neutral Disagree Strongly Disagree1st Question 4 16 5 2 3

2nd Question 3 5 2 4 16

3rd Question 2 3 16 4 5

4th Question 16 5 4 3 3

Example: let us assume that the scaled question has been answered by a selected sample of 30 students. The student’s responses per question are presented in table-1. The response rate in this case is 100% because all of the students have participated in answering the survey questions

1st Question Numerical Value Responses Strongly Agree 1 4

Agree 2 16Neutral 3 5

Disagree 4 2Strongly Disagree 5 3

Table-1

Table-2

Table-2 is extracted from table-1 to show how the students feel about the 1st question


1 2 3 4 50

2

4

6

8

10

12

14

16

18

4

16

5

23

Question-1 Options relative to Responses

Options of Question-1

Re

sp

on

se

s

It is clearly noticed that nearly half of students have chose option-2, which means that they have no clear understanding of the concept behind LAB experiments, while the others indicate that there some degree of understanding. The variation of responses is attributed to two factors; 1st factor: no adequate preparation nor pre-lab questions were solved. 2nd factor: tutor shall dedicate half an hour time to explain the theoretical concepts prior each LAB

Fig.1 The chart investigates the correlation of student responses on question-1

Discussion


Sample Questionnaire Example

References and Further Reading

http://betterevaluation.org/evaluation-options/stratifiedrandom

https://www.roadsafetyevaluation.com/evaluationtopics/researchmethods/sampling.html

http://www.fao.org/docrep/004/y2790e/y2790e05.htm

http://www.statisticalconsultants.co.nz/weeklyfeatures/WF7.html

http://www.britannica.com/EBchecked/topic/564172/statistics

http://home.ubalt.edu/ntsbarsh/stat-data/Surveys.htm#rssm

http://www.statpac.com/surveys/sampling.htm

http://www.sciencebuddies.org/science-fair-projects/project_ideas/Soc_survey.shtml

http://www.socialresearchmethods.net/kb/survey.php

http://www.accesscable.net/~infopoll/tips.htm

http://www.surveysystem.com/sdesign.htm

http://asq.org/learn-about-quality/data-collection-analysis-tools/overview/survey.html

http://www.sciencebuddies.org/science-fair-projects/project_ideas/Soc_survey.shtml


http://web.njcu.edu/sites/faculty/kivy/Uploads/survey_research.pdf

References and Further Reading

https://www.roadsafetyevaluation.com/evaluationtopics/info/tecniques-for-selecting-samples.pdf

http://www.aapor.org/AM/Template.cfm?Section=Cell_Phone_Task_Force_Report&Template=/CM/ContentDisplay.cfm&ContentID=3189

Robert M. (2009). Survey methodology. 2nd edition. USA:Wiley. 488 pages

http://www.who.int/chp/steps/Part4.pdf#page=6&zoom=130,0,820


Principles of survey research

Education

survey data

survey step

large survey

survey design

survey types of data

survey process page

list of survey questions

final stage of survey