ED 512 Quantitative Research Methods

 

Course Home || Syllabus || Final Project || Descriptive/survey research ||

 

Selecting a topic

 

Descriptive research

 

Experimental Research

 

Causal Comparative

 

Descriptive Statistics

 

Inferential Statistics

 

 

The survey is a research technique in which data are gathered by asking questions of respondents.

Before you read the notes below please respond to the following question:

Why is survey research the same as descriptive research?

Types of surveys
Before initiating survey research, the investigator must determine the format that is most appropriate for the proposed investigation. Surveys are classified according to their focus and scope.

Surveys classified according to focus and scope
A survey that covers the entire population of interest is referred to as census. An example of census is the U.S. census undertaken by government every ten years. A survey that studies a portion of the population is known as a sample survey.

Census of tangibles
Imagine a principal wants to know how many desks, how many children ride the bus, how many teachers have a masters’ degree, a simple count will provide the information. Such a study would cover an entire population of a school, hence the descriptive term census. Tangibles are well defined and unambiguous variables; the strength of census of tangibles is its irrefutability.

Census of intangibles
Suppose the principal now seeks information about parents’ attitudes toward school. This task deals with intangible variables or constructs that are not directly observable but must be inferred from indirect measures. To design a tool to investigate parents’ attitudes requires expertise and knowledge of attitude indicators. Administering the scale to all parents in a school system would represent a census of intangibles.
Does census of intangibles remind you of measurement chapter? If it does it is because you are dealing with the phenomenon – measuring attitudes. You want an instrument or tool designed to ask all the right questions. Perhaps you will choose a Likert Scale.

A sample of tangibles
Seeking information from a large group can be very expensive so researchers use sampling techniques and use the information they collect form the sample to make inferences about the population as a whole. Example of sample survey is the Coleman Report (1966).

A sample survey of intangibles
Public opinion pools are examples of studies measuring intangible constructs. Opinion is not directly observable and is inferred from responses made by subjects to questionnaires or interviews. The Florida presidential election debacle can be attributed to measuring an intangible using a sample to predict how the state will vote. The newscasters reported that Florida had voted for Gore whereas what turned to be the fact is that most votes belonged to Bush.

Surveys classified according to the time dimension
Two types of surveys are classified according to the time of data collection; longitudinal and cross- sectional surveys.

Longitudinal surveys
They gather information at different points in time in order to study changes over extended periods of time. For example, studying the development of quantitative reasoning in elementary school children would select a sample of first graders and administers a measure of quantitative reasoning. This same group would be followed through successive grade levels and tested each year to assess how quantitative reasoning abilities develop over time.

Panel studies
The same subjects are surveyed at different times over an extended period of time. See the above example.

Trend studies
Different people from the same general population thus each year from 2000 6th graders are surveyed over at different times perhaps 2001, 2001, 2005 to understand the trend of this particular population.

Cohort studies
A specific population is followed over a length of time. For example, school graduating class of 2003 might be followed over a length of time. From a list of 2003 graduates, a random sample is drawn at different points in time and data are collected from that sample. Thus the population remains the same during the study, but the individuals surveyed are different each time.

Cross-sectional surveys
These surveys are used to study a sample of a population at a single point in time. A cross sectional study would compare quantitative reasoning of a sample of students from grades 1, 2, 3, 4, and 5. The cross sectional survey is the method of choice if you want to gather the data at one point in time.

The Survey Technique: Six basic steps are involved in survey research
1. Planning: is survey the appropriate means for data gathering?
2. Define the population: to whom will you distribute the survey? Construct a complete list of all individuals in the population.
3. Select a representative sample
4. Construct an instrument: questionnaires or interviews (measurement section)
5. Field test the instrument.
6. Process the data.

Data gathering techniques:
1. Personal interview – flexible, questions can be repeated, or their meaning explained, personal contact increases the likelihood that the individual will respond, the main disadvantage is the expense, and a possibility of interviewer bias

2. Telephone interview: compares favorably with face to face interviews, greater freeing of anonymity, takes skill to carry out a telephone interview so that valid results are obtained, household may not have telephones or have unlisted their numbers, although this can be solved using the random digit dialing

3. Mailed questionnaire: guarantees confidentiality or anonymity, disadvantage is that subjects may misinterpret the questions thus a complex survey is completed by people with considerable education. Questionnaires have a low return rate, furthermore non-response rate may not be randomly distributed in the population making generalizability questionable. Email surveys are becoming more popular. They have prompter returns and have lower item non-response and more complete answers to open ended questions. Plan for a prior email notification and you are likely to have more response, try to address each individual person instead of using a mailing list. Internet surveys the advantage of these is a similar format for all respondents, and are easier for the respondent to navigate – e.g. Zoomerang.

4. Directly administered questionnaire: group assembled in a place – college classroom. High response rate is one big advantage and the ease in reaching the participants. The researcher can provide assistance. The main disadvantage is the researcher is restricted in terms of where and when the questionnaire can be administered.

Please respond to these questions and bring your responses to class on July 25.


Survey or Descriptive research questions.
1. Suggest a research question that can best be answered by means of a survey. Write one open-ended and one closed-end items, and one likert scale items that would provide data relevant to the question.

1. What data gathering technique would you use for each of the following surveys?
a) A survey of the sample of high school teachers throughout the state concerning the use of mandatory competency exams for high school graduation.
b) A survey of the opinions of people in a major metropolitan area on the way juveniles who commit violent crimes are presently handled in the state’s court system?
c) A freshman survey of certain non-cognitive characteristics that might relate to academic achievement in the first year of college.

3. Turn to page 166 of the 8th edition text. There is a sample questionnaireitems in a survey of high school teachers

a) Examine the Demographics, Checklist, Likert Scale, and Free response sections of the questionnaire. These sections are tools to measure tangible or intangible variables. Distinguish the intangible and the intangible variables.

b) What are the challenges of measuring a) the intangibles and b) the tangibles? Comment on the challenges.

4. Suggest a research question that can best be answered by means of a survey. Write one open-ended and one closed-end question.

5. On the basis of the time of data collection, classify each of the following surveys:

1. Terman's study of adults who were intellectually gifted as children

2. A comparison of math achievement in public middle schools in the United States in 1990, 1995, and 2000

3. A follow-up of the 1995 graduates of the Indiana University School of Business

4. A survey of reading achievement at different grade levels in a school system in 2000.