RESEARCHING THE REAL WORLD



MAIN MENU

Basics

Orientation Observation In-depth interviews Document analysis and semiology Conversation and discourse analysis Secondary Data Surveys Experiments Ethics Research outcomes
Conclusion

References

Activities

Social Research Glossary

About Researching the Real World

Search

Contact

© Lee Harvey 2012–2019

Page updated 25 January, 2019

Citation reference: Harvey, L., 2012–2019, Researching the Real World, available at qualityresearchinternational.com/methodology
All rights belong to author.


 

A Guide to Methodology

8. Surveys

8.1 Introduction to surveys
8.2 Methodological approaches
8.3 Doing survey research

8.3.1 Aims and purpose
8.3.2 Background to the research
8.3.3 Feasibility
8.3.4 Hypotheses
8.3.5 Operationalisation
8.3.6 How will data be collected and what are the key relationships?
8.3.7 Designing the research instrument
8.3.8 Pilot survey
8.3.9 Sampling
8.3.10 Questionnaire distribution and interviewing
8.3.11 Coding data
8.3.12 Analysis

8.3.12.1 Response rate
8.3.12.2 Frequency tables
8.3.12.3 Graphical representation
8.3.12.4 Measures of central tendency (averages)
8.3.12.5 Levels of measurement
8.3.12.6 Crosstabulation
8.3.12.7 Measures of dispersion
8.3.12.8 Generalising from samples
8.3.12.9 Dealing with sampling error
8.3.12.10 Confidence limits
8.3.12.11 Statistical significance
8.3.12.12 Association
8.3.12.13 Summary of significance testing and association: an example

8.3.13 Hypothesis testing
8.3.14 Significance tests
8.3.15 Report writing

8.4 Summary and conclusion

8.3.12.2 Frequency tables
usually, the first stage of analysis is to produce frequency tables for each variable. Frequency tables provide a count of the number in the sample who fall into each value of the variable. They can also be drawn up to show the appropriate percentages. In effect, constrcuting a frequency table involves counting the numbers of times each code occurs in a column of a data file.

Consider variable 1 in the CASE STUDY data file, 'Had you previously heard of Clause 28?'. Reading down the column there are fifty '1's, ninety-nine '2's and two '9's. This can be presented as a frequency table as in Table 8.3.12.1 This can, of course, be done quickly if you use a computer software package that is designed for this purpose.

 

Table 8.3.12.1: V1. Had you previously heard of Clause 28?

Value
Label

Value
Frequency
Percentage

Valid
Percentage

Yes
1
50
33.1
33.6
No
2
99
65.6
66.4
Missing
9
2
1.3
Total
151
100.0
100.0

 

Table 8.3.12.1 provides a model that you would be advised to follow when creating frequency tables. Note that the table has a heading with the variable number and name, and the value labels are on the left followed by the codes used to represent them in the data file. The third column is the actual frequencies (the number of time each code appears). The fourth column gives the percentage that each frequency is of the whole sample. The fifth column is the percentage adjusted for missing values, this is called the valid percentage.

So, in the example, 50 people said 'Yes' (Value code 1) they had previously heard of Clause 28, 99 said 'No' (value code 2), and there were 2 missing cases (value code 9).

In percentage terms, 33.1% said 'Yes', (that is, (50/151) x 100 = .331 x 100 = 33.1%) 65.6% said 'No' and there was no information on 1.3% of the sample.

If the missing values are excluded then the 'valid percentage' (that is, calculated on a reduced sample of 149) is 33.6% 'Yes' and 66.4% 'No'.

Normally, you would construct the frequency tables for each of the variables. This can be a time-consuming process if you do not have a computer but you usually need this basic data for two reasons.

First, it provides a detailed summary of the responses so that you have a complete description of the sample.

Second, it acts as another check on the data file. If you come across a code that is not possible (for example, if there was a 7 in the column for variable 3) then you must go back to the original schedule and find out what the proper value should be. You do this by finding the questionnaire with the same identification number as the one recorded at the start of the row that contains the incorrect code. Look up the appropriate variable on the original questionnaire or schedule and then change the incorrect code in the data file, inserting the correct one. You will then need to correct your frequency table if it has already been compiled.

Activity 8.3.12.1
Using the data set for the CASE STUDY Attitudes towards homosexuality: Data
1. Compile a frequency table for variable 3 similar to Table 8.3.12.1. What percentage of the sample agree with Clause 28? Does this suggest that the population being surveyed support government initiatives?
2. Refer to the hypotheses for the CASE STUDY, what other frequency tables would you need to assess all parts of hypothesis 2? Compile the frequency tables necessary. Is hypothesis 2 confirmed? What do you conclude from the results?

To test sub-hypothesis 1 from the CASE STUDY on attitudes towards homosexuality it is necessary to analyse the respondents' attitudes about the following:

  • whether homosexuality should be legal (variable 23 (V23));
  • what the consenting age for adults should be (V21 and V22);
  • is the idea of gay and lesbian families spported (V28 and V29);
  • and are public homosexual affectionate acts accepted(V24 and V25)?

Frequency tables can provide some initial insights. Table 8.3.12.2 is the frequency table for V23.

 

Table 8.3.12.2: V23 'Should homosexuality be made illegal?'

Value
Label

Value
Frequency
Percentage

Valid
Percentage

Yes
1
44
29.1
30.8
No
2
99
65.6
69.2
Missing
9
8
5.3
Total
151
100.0
100.0

It shows that a large minority, 30.8%, of those who answered the question (in 1990) thought that homosexuality should be made illegal.

 

Table 8.3.12.3: V22 What should the consenting age be for men?

Value
Frequency
Percentage

Valid
Percentage

Cumulative
Percentage
12
1
0.7
0.9
0.9
14
2
0.7
0.9
1.8
15
1
0.7
0.9
2.7
16
28
18.5
25.2
27.9
17
2
1.3
1.8
29.7
18
33
21.9
29.7
59.5
19
1
0.7
0.9
60.4
20
8
5.3
7.2
67.6
21
26
17.2
23.4
91.0
25
2
1.3
1.8
92.8
30
2
1.3
1.8
94.6
60
1
0.7
0.9
95.5
78
1
0.7
0.9
96.4
80
1
0.7
0.9
97.3
98
3
2.0
2.7
100.0
Missing
40
26.5
Total
151
100.0
100.0

Note: there are no value labels for Table 8.3.12.3 as the values themselves refer to ages in years, except the value 9 which is the code for missing values.

The frequency table for V22 suggests that the sample do not support the sub-hypothesis 1b ('the age limit for consenting adults should be raised').

Of those who answered the question, 67.6% think the age of consent for gays should be under 21 and only 9.0% think it should be over 21. However, the values in Table 8.3.12.3 range from 12 to 98. The six high ages of 60 or more, suggest that the respondents are opposed to homosexuality, and this may well be the case for most of the 40 missing values. Thus, there are 30.6% of the sample who have given no answer or a very high one. This is virtually the same percentage as think homosexuality should be illegal.

So, in this case, a more reasonable estimate of percentage who favoured a lower age limit would be one based on the entire sample, whether they answered the question or not. A total of 75 (49.7%) of the whole sample of 151 indicated that the age limit should be less than 21 (see the cumulative percentage column).

The cumulative percentage adds up the valid percentages as you go down the table. In the example, 0.9% think the consenting age should be 12, 1.8% think it should be 14 or less, 2.7% think it should be 15 or less, and so on. Thus 59.5% of the valid responses think it should be 18 or less.

 

See also DATA ANALYSIS: A BRIEF INTRODUCTION Section 5 (Downloads a Word document into your downloads folder)

 

Next 8.3.12.3 Graphical representation

Top