Thứ Sáu, ngày 12 tháng 3 năm 2010

Chapter 2 - Analysing data with computers

First steps with SPSS 10 for Windows

Since the different kinds of statistics to be described in this book will be carried out with one of the, if not the, most widely used and comprehensive statistical programs in the social sciences, SPSS, we will begin by outlining what this entails. The abbreviation SPSS stands for Statistical Package for the Social Sciences. This package of programs is available for both personal and mainframe (or multi-user) computers. These programs are being continuously updated and so there are various versions in existence.

Currently there are two main kinds of operating system for computers. The traditional system, still employed by mainframe (or multi-user) computers, requires commands and names to be typed in. The more recent system uses menus and dialog boxes from which these commands and names can be selected by keys or a mouse, although commands can also be typed in. This latter system was originally developed for Macintosh personal computers and is now available for a Windows environment on IBM-compatible personal computers with a 386 or higher processor. The version for Windows 3.1 is known as SPSS Release 6, which we have outlined in an earlier book (Bryman and Cramer, 1997). After the introduction of Windows 95, a new release of SPSS (Release 7) was designed to run on this operating system. The latest version for Windows 95/98 is Release 10. This book describes the use of Release 10, which we shall refer to as SPSS for short unless otherwise indicated. The use of Releases 8 and 9 was described in the previous edition of this book (Bryman and Cramer, 1999).

The great advantage of using a package like SPSS is that it enables you to score and to analyse quantitative data very quickly and in many different ways once you have learned how. In other words, it will help you to eliminate those long hours spent working out scores, carrying out involved calculations, and making those inevitable mistakes that so frequently occur while doing this. It will also provide you with the opportunity to use more complicated and often more appropriate statistical techniques which you would not have dreamed of attempting otherwise.

There is, of course, what may seem to be a strong initial disadvantage in using computer programs to analyse data, which is that you will have to learn how to run these programs. The time spent doing this, however, will be much less than doing these same calculations by hand. In addition, you will have picked up some knowledge which should be of value to you in a world where the use of computers is fast becoming increasingly common. The ability to do things quickly and with little effort is also much more fun and often easier than you might at first imagine.

When mastering a new skill, like SPSS, it is inevitable that you will make mistakes which can be frustrating and off-putting. While this is something we all do, it may seem that we make many more mistakes when learning to use a computer than we do carrying out other activities. The reason for this is that programs require instructions to be given in a very precise form and usually in a particular sequence in order for them to work. This precision may be less obvious or true of other everyday things that we do. It is worth remembering, however, that these errors will not harm the computer or its program in any way.

In order to make as few mistakes as possible, it is important at this stage to follow precisely the instructions laid down for the examples in this and subsequent chapters. Although ‘bugs’ do sometimes occur, errors are usually the result of something you have done and not the fault of the machine or the program. The program will tell you what the error is if there is something wrong with the form of the instructions you have given it, but not if you have told it to add up the wrong set of numbers. In other words, it questions the presentation but not the objectives of the instructions.
THE DATA FILE

Before you can analyse data, you need to create a file which holds them. To illustrate the way in which these files are produced, we shall use an imaginary set of data from a questionnaire study which is referred to as the Job Survey. The data relating to this study derive from two sources: a questionnaire study of employees who answer questions about themselves and a questionnaire study of their supervisors who answer questions relating to each of the employees. The questions asked are shown in Appendix 2.1 at the end of this chapter, while the coding of the information or data collected is presented in Table 2.1. The cases consist of people, traditionally called respondents by sociologists and subjects by psychologists whose preferred term now is participants. Although questionnaire data have been used as an example, it should be recognised that SPSS and the data analysis procedures described in this book may be used with other forms of quantitative data, such as official statistics or observational measures.

As the data set is relatively large, it may be more convenient to let skilled personnel enter the data into a file for you if you have access to such a service. If you do, they may enter it into what is called a simple text or ASCII file. ASCII stands for American Standard Code for Information Interchange and is widely used for transferring information from one computer to another. You then read this data file into SPSS. If you do not have access to such a service or if the data set is small, then it may be easier for you to enter the data directly into an SPSS window called Data Editor. Both these procedures are described later in this chapter.

With a simple text file, the data are put into a space which consists of a large number of rows, comprising a maximum of eighty columns in many computers. Each column in a row can only take one character such as a single digit. The data for the same variable are always placed in the same column(s) in a row and a row always contains the data of the same object of analysis or case. Cases are often people, but can be any unit of interest such as families, schools, hospitals, regions or nations.

Since it is easier to analyse data consisting of numbers rather than a mixture of numbers and other characters such as alphabetic letters, all of the variables or answers in the Job Survey have been coded as numbers. So, for instance, each of the five possible answers to the first question has been given a number varying from 1 to 5. If the respondent has put a tick against White/European, then this response is coded as 1. (Although the use of these categories may be questioned, as may many of the concepts in the social sciences, this kind of information is sometimes collected in surveys and is used here as an example of a categorical (nominal) variable. We shall shorten the name of the first category to ‘white’ throughout the book to simplify matters.) It is preferable in designing questionnaires that, wherever possible, numbers should be clearly assigned to particular answers so that little else needs to be done to the data before they are typed in by someone else. Before multiple copies of the questionnaire are made, it is always worth checking with the person who types in this information that this has been adequately done.

It is also important to reserve a number for missing data, such as a failure to give a clear and unambiguous response, since we need to record this information. Numbers which represent real or non-missing data should not be used to code missing values. Thus, for example, since the answers to the first question on ethnic group in the Job Survey are coded 1 to 5, it is necessary to use some other number to identify a missing response. In this survey all missing data except that for absenteeism have been coded as zero since this value cannot be confused with the way that non-missing data are represented. Because some employees have not been absent from work (that is, zero days), missing data for absenteeism could not be coded as ‘0’. Instead, it is indicated by ‘99’ since no employee has been away that long. Sometimes it might be necessary to distinguish different kinds of missing data, such as a ‘Don’t know’ response from a ‘Does not apply’ one, in which case these two answers would be represented by different numbers.

It is advisable to give each participant an identifying number to be able to refer to them if necessary. This number should be placed in the first few columns of each row or line. Since there are seventy participants, only columns 1 and 2 need to be used for this purpose. If there were 100 participants, then the first three columns would be required to record this information as the largest number consists of three digits. One empty or blank space will be left between columns containing data for different variables to make the file easier to read, although it is not necessary to do this.

Since all the data for one participant can be fitted on to one line using this fixed format, only one line needs to be reserved for each participant in this instance, and the data for the next participant can be put into the second line. If more than one line were required to record all the data for one participant, then you would use as many subsequent rows as were needed to do so. In this case, it may also be worth giving each of the lines of data for a particular participant an identifying number to help you read the information more readily, so that the first line would be coded 1, the second 2, and so on. Each line or row of data for a participant is known as a record in SPSS.

The first variable in our survey and our data file refers to the racial or ethnic origin of our respondents. Since this can only take one of six values (if we include the possibility that they might not have answered this question), then these data can be put into one column. If we leave a space between the participant’s two-digit identification number and the one-digit number representing their ethnic group, then the data for this latter variable will be placed in column 4. Since the second variable of gender can also be coded as a single digit number, this information is placed in column 6. The third variable of current gross annual income, however, requires five columns of space since all participants earned more than £10,000 but less than £100,000, and so this variable occupies columns 8 to 12 inclusively (please note that the comma and pound sign should not be included when entering the data).

A full listing of the variables and the columns they occupy is presented in Table 2.2. The data file is named jsr.dat which is an abbreviation of ‘ job survey raw data’. Since SPSS accepts letters written in capitals, or upper case (for example, JSR.DAT ) and small, or lower case (for example, jsr.dat), lower-case letters will be used to make typing easier for you. Restrictions and conventions on the form of names will be described later in this chapter.

Table 2.2 The SPSS names and location of the Job Survey variables

Variable name

SPSS name

Column location

Identification number

id

1–2

Ethnic group

ethnicgp

4

Gender

gender

6

Gross annual income

income

8–12

Age

age

14–15

Years worked

years

17–18

Organisational commitment

commit

20

Job-satisfaction scale

Item 1

satisl

22

Item 2

satis2

24

Item 3

satis3

26

Item 4

satis4

28

Job-autonomy scale

Item 1

autonoml

30

Item 2

autonom2

32

Item 3

autonom3

34

Item 4

autonom4

36

Job-routine scale

Item 1

routinel

38

Item 2

routine2

40

Item 3

routine3

42

Item 4

routine4

44

Attendance at meeting

attend

46

Rated skill

skill

48

Rated productivity

prody

50

Rated quality

qual

52

Absenteeism

absence

54–55




GAINING ACCESS TO SPSS

To use SPSS, it is necessary to have access to it via a personal computer. A personal computer consists of a keyboard on which you type in your instructions, a mouse which provides an alternative way of moving about the screen and selecting instructions, and usually a video display unit (VDU) or television-like screen which displays information. While the amount of information shown at any one moment on the screen is necessarily limited, further information can be brought into view with the appropriate use of the keys or the mouse. A personal computer also usually has a printer which can be used to print out information stored in the computer and can be used to print out a record of what you have done. Keyboards are used to type or put in (hence the term input) the data that you want to analyse and also the names of variables and files you have created.

The Windows system allows commands to be selected from words or icons presented as a menu in a window on the screen. Commands can usually be selected by moving a pointer called a cursor on to them with either the keys or, more normally, the mouse, and then pressing the Return key or the left button on the mouse, or in Windows 95/98 by simply selecting the next option. Choosing options with the mouse is generally easier than doing this with keys since it simply involves moving the mouse appropriately. With keys, however, some options are chosen by pressing the relevant cursor keys while others are selected by pressing up to two keys other than the cursor keys. The cursor keys are usually on the right hand side of the keyboard and have arrows on them pointing in the direction in which the cursor is to be moved. You may prefer to use the mouse for some operations and the keys for others.

To access SPSS in the windows environment, select the button or icon at the bottom of the screen which presents the first column or menu on the left in Box 2.1.

Select Programs on this menu which displays the second menu (columns 2 and 3) in Box 2.1. Note that on many computers fewer programs will be listed than shown here.

Select SPSS for Windows from this menu, which opens the third menu (at the bottom of column 3) in Box 2.1.

Select SPSS 10.0 for Windows which produces the Data Editor window in Box 2.2. You can prevent the SPSS for Windows dialog box superimposed on the

Box 2.1 Windows 95/98 opening window

Box 2.2 SPSS Data Editor

Data Editor being shown on your own computer the next time you access SPSS if you select the Don’t show this dialog in the future check box near the bottom of the dialog box. Select Cancel to remove the dialog to enter data into the cells of the Data Editor. Listed at the top of this window are the names of various procedures such as

Box 2.3 Data option drop-down menu

File, Edit and so on. To see what these procedures are, we simply move the cursor to a particular option and press the left button on the mouse once. A drop-down menu will then appear, as shown in Box 2.3 where the Data option has been chosen. To see the other options, simply move the cursor to that option.

The ellipsis, or three dots, after an option term (…) on a drop-down menu, such as on the Select Cases… option, signifies a dialog box will appear when this option is chosen. If we select this option, for example, the Select Cases dialog box displayed in Box 2.4 will appear when data have already been entered into the Data Editor. To cancel a dialog box, select the Cancel button in the dialog box. A right-facing arrowhead ► after an option term such as on the Merge Files option, on the other hand, indicates that a further submenu will appear to the right of the drop-down as shown in Box 2.8. An option with neither of these signs means that there are no further drop-down menus to select.

Below these options is a toolbar with buttons on it. These buttons enable you to carry out procedures directly without having to go to the options and select the appropriate procedure. The functions that the buttons carry out are displayed in a small yellow box near them and in the bottom line of the window. So, for example, the first button says Open File. You can add further buttons to the toolbar. The Help system described on pp. 33–4 gives instructions on how to do this.

Box 2.4 Select Cases dialog box

ENTERING AND EDITING DATA IN DATA EDITOR

The easiest way to enter data in SPSS yourself is to type it directly into the matrix of columns and numbered rows in the Data Editor window shown in Box 2.2. Note that in this case a column can hold more than one digit. Initially the cursor will be in the cell in the first row of the first column. The frame of this cell will be shown in bold to denote that it is the active cell. To enter a value in any one cell, make that cell active by moving to it with either the cursor keys or the mouse, type in the value and then move to the next cell into which you want to put a value. Columns are consecutively numbered once you enter a value. So if you enter a value in the fifth column the first five columns will be labelled var00001 to var00005. To change any value already entered, move to the cell containing that value, type in the new value and move to another cell. If you want to leave a cell empty delete the entry with the Backspace or Delete key and move to another cell, when a full stop (.) will be left denoting a missing value.
NAMING VARIABLES IN DATA EDITOR

To name a variable in Data Editor, select Variable View near the bottom of the window. Select the appropriate row under the Name column and type in the name (for example, ethnicgp in the first row as shown in Box 2.5).
SPSS NAMES

Variable and file names in SPSS have to meet certain specifications. They must be no longer than eight characters and must begin with an alphabetic character (A–Z). The remaining characters can be any letter, number, period, @ (at), $ (dollar) or _ (underscore). Blank spaces are not allowed and they cannot end with a period and, preferably, not with an underscore. In addition certain words, known as keywords, cannot be used because they can only be interpreted as commands by SPSS. They include words such as add, and, any, or and to, to give but a few examples. If you accidentally use a prohibited keyword as a name, you will be told this is invalid when you try to run this procedure by selecting the OK button. No keyword contains numbers so you can be certain that names which include numbers will always be recognised as such. It is important to remember that the same name cannot be used for different variables or files. Thus, for example, you could not use the name satis to refer to all four of the questions which measure job satisfaction. You would need to distinguish them in some way, such as calling the answer to the first question satisl, the answer to the second one satis2, and so on. The SPSS names given to the variables in the Job Survey are presented in Table 2.2.

Box 2.5 Variable View window of Data Editor
DEFINING OTHER ASPECTS OF VARIABLES IN DATA EDITOR

We can define nine other aspects of variables when naming them. These aspects are listed just above the data matrix and range from Type on the left to Measure on the right. You may not see all these nine aspects at once. You can change the width of each column by selecting the line next to the name of the column whose width you want to change and moving the column line to the desired position, as we have done in Box 2.5.

The pre-selected settings for these aspects are shown and are known as the default options. If we wish to change any of these settings, we select the appropriate row and column to make the desired change. In general and for our purposes the most important of these other aspects is Missing values.
DEFINING MISSING VALUES

In the Job Survey, we have missing values for income (cases 12 and 21), age (case 45), satis1 (cases 1 and 2), satis2 (case 2), prody (case 1) and absence (case 31). So we have to specify the appropriate missing values for these variables, which are 0 for the first five (income, age, satis1, satis2 and prody) and 99 for the sixth variable called absence. We do this by selecting the appropriate row of the Missing column in the Variable View of the Data Editor and then selecting the ellipsis or three dots in that cell. This opens the Missing Values dialog box shown in Box 2.6. In our case we select Discrete missing values, type the appropriate value in the first box and then select OK. If we type in 0, then None in that row of the Missing column of Variable View will be replaced with 0 as shown in Box 2.5.

If someone else is entering the data, we need to let them know how missing data for any of the variables are to be coded. We could enter this code here to remind us what it is. Thus, in this example, missing data for all variables other than absence has been defined as 0.
DEFINING DECIMAL PLACES

The default number of decimal places is two. For most purposes it is easier to code all variables as numbers, which we have done for the Job Survey. Since all our values are whole numbers we could change 2 to 0 in the Decimals column of the Variable View window. To do this, we select each row in turn in this column, and select the downwards button to give 0.
DEFINING VARIABLE AND VALUE LABELS

SPSS variable names are restricted to eight characters, which usually means that they have to be abbreviated, making their meaning less clear. Using this option, variable labels can be created which will be displayed on the output. These variable labels can be very long, although most output will not present very long labels. For example, the SPSS variable name ethnicgp could be labelled ethnic

Box 2.6 Missing values dialog box

group. To do this, we type in this name for the first row of the Labels column in the Variable View window. If we do this, then the extended variable name will be presented first in the section of dialog boxes where the variable names are listed (for example, Box 2.12) followed by the abbreviated name in brackets. We have used abbreviated names as these are used in the other sections of the dialog boxes and are generally less cumbersome.

We could also label the values of a variable by selecting the appropriate row in the Values column (for example, ethnicgp) and then selecting the ellipsis, or three dots, in that cell. This opens the Value Labels dialog box shown in Box 2.7. Type in the value (for example, 1) in the box entitled Value:, the label (for example, white) in the box entitled Value Label: and select Add. The value labels for the five ethnic groups in the Job Survey are presented in Box 2.7. Value labels can be up to sixty characters long, although most output will not show labels this long. To remove a label we first select it and then Remove. To change a label, we first select it, make the desired changes and then select Change. When we have what we want, we select OK to close the Value Labels dialog box.
DEFINING COLUMN FORMAT AND ALIGNMENT

It is unlikely that you would wish to change the width of the column in Data Editor, but if you do, select the appropriate row in the Columns column and then select the upwards or downwards button to enter your desired value. If you wish to alter the alignment of data within a column, select the appropriate row in the Align column, select the downwards arrow and then select one of the other two options.
Defining consecutive variables simultaneously

If you want to define consecutive variables simultaneously with the same format (such as sat1 to routine4), define the first variable (satis1), copy that row,

Box 2.7 Value Labels dialog box
highlight the subsequent rows to be defined ( 9 to 19), select Paste or Paste Variables… and re-name variables as appropriate.
SAVING DATA IN DATA EDITOR

When we want to leave SPSS or to work on another data set in the same session, these data and any changes we have made to them will be lost unless we save them as a file. We could save this file on to the hard disk of the computer. However, if others use the computer, they may delete our files. Even if no one else is likely to use the computer, it is necessary to make a back-up copy of our files on one or more formatted floppy disks in case we should lose them. The floppy disk is inserted into a slot called a drive.

To be able to retrieve a file, we need to give it a name. This name can consist of a prefix or stem of up to eight characters followed by a full stop and a suffix or extension of three characters. The stem name usually refers to the content of the file (such as jsr in the case of our Job Survey raw data file) while the extension

Box 2.8 File drop-down menu

Box 2.9 Save Data As dialog box

name refers to the type of file. The default extension name for files in Data Editor is sav. Thus, we could call our data file jsr.sav. Extensions are now usually displayed as icons and not names (as shown in Box 2.10).

We shall generally use a particular notation throughout this book as shorthand to describe the steps involved in any application. The selection of a step or option will be indicated with a right facing-arrow ➔ pointing to the term(s) on the menu or dialog box to be chosen. Any explanations will be placed in square parentheses after the option shown. Steps in a dialog box or a subdialog box (which is a box which can only be accessed after the initial dialog box has been opened) will begin on a new line. The sequence will be indented. Thus, the notation for saving this file on a floppy disk in Drive A is:

➔File [shown in Box 2.8] ➔Save As… [opens Save Data As dialog box shown in Box 2.9] type a:\jsr.sav in box beside File name: ➔Save
RETRIEVING A SAVED DATA EDITOR FILE

To retrieve this file at a later stage when it is no longer the current file, use the following procedure:

➔File ➔Open [opens Open File dialog box shown in Box 2.10] type a:\jsr.sav in box beside File name: ➔Open

READING AN ASCII DATA FILE IN DATA EDITOR

If the data have been saved on a floppy disk as an ASCII file called jsr.dat, then carry out the following sequence to put it in Data Editor:

➔File ➔Read Text Data… [opens Open File dialog box shown in Box 2.10] type a:\jsr.dat in box beside File name: ➔Open [opens Text Import Wizard – Step 1 of 6 dialog box shown in Box 2.11]

➔Next > [opens Text Import Wizard – Step 2 of 6 dialog box]

➔Fixed width ➔Next > [opens Text Import Wizard – Step 3 of 6 dialog box]

➔Next > [opens Text Import Wizard – Step 4 of 6 dialog box]

➔Next > [opens Text Import Wizard – Step 5 of 6 dialog box]

➔Next > [opens Text Import Wizard – Step 6 of 6 dialog box]

➔Finish [closes Text Import Wizard – Step 6 of 6 dialog box and displays data in Data View]

➔Variable View [to define data]
STATISTICAL PROCEDURES

After entering the data set in Data Editor, we are now ready to analyse it. The rest of the book describes numerous ways in which you can do this. To show you how this is generally done, we shall ask SPSS to calculate the average or mean age of the sample. This can be done with a number of SPSS commands, but we shall use the one called Descriptive Statistics. This works out a number of other descriptive statistics as well. The procedure for doing this is:

➔Analyze ➔Descriptive Statistics ➔Descriptives... [opens Descriptives dialog box shown in Box 2.12]

➔variable [for example, age; note variables are listed in their order in Data Editor] ➔►button [puts the selected variable in box under Variable[s]:] ➔OK

The output for this procedure is displayed in the Viewer window as shown in Table 2.3. The mean age is 39.19. The other descriptive statistics provided by default are the standard deviation (see Chapter 5), the minimum age, the maximum age, and the number of cases ( N) on which this information is based. If we look at the ages in our Job Survey data, we can confirm that the minimum age is indeed 18 (for the first case) while the maximum is 63 (for case number 5). We should also notice that the age of one of our participants (case number 45) is missing, making the number of cases which provide valid data for this variable 69 and not 70.

As shown in Table 2.3 the output in the Viewer window will always be preceded by the name of the statistical procedure. In this case, it is Descriptives. These titles will be omitted from subsequent presentations of output to save space.

If we wanted just the mean age and not the other statistics, we could do this

Box 2.10 Open File dialog box

Box 2.11 Text Import Wizard - Step 1 of 6 dialog box

Box 2.12 Descriptives dialog box

Table 2.3 Default Descriptives output







as follows. Select Options… in the Descriptives dialog box to open the Descriptives: Options subdialog box shown in Box 2.13. Then de-select Std. deviation, Minimum and Maximum by moving the cursor on to them and pressing the left button. The output for this procedure is presented in Table 2.4. If you want to revert to the Data Editor, either select jsr – S… at the bottom of the screen or select the Window option and then 1 jsr – SPSS Data Editor from the drop-down menu. If an SPSS operation has been started but not completed (in that all dialog boxes concerned with that operation have not been closed), scrolling through the Viewer will not be possible.
SAVING AND PRINTING OUTPUT

To print the contents of any window, enter that window and then execute the following sequence:

➔File ➔Print… ➔OK

If you want to store the contents of any window on a floppy disk, then carry out the following steps:
Box 2.13 Descriptives: Options subdialog box

Table 2.4 Mean Descriptive output

Descriptive Statistics




N


Mean

AGE


69


39.19

Valid N (listwise)


69




➔File ➔Save As ➔window [opens Save As dialog box]

➔type the drive and file name in the box beside File name: [for example, a:\jsr.spo] ➔Save

The default extension name for output files is spo which is short for spss output file. You can edit output files before saving them. For example, you may wish to delete certain analyses or type in some further explanation.
HELP SYSTEM

SPSS has a Help system which you may like to use to avoid having to refer to a book like this one or to find out more about the program. As this system is meant to be self-explanatory you should be able to learn to use it yourself after a little experience. To find help on a topic such as file, carry out the following procedure:
Box 2.14 Help Topics Content dialog box

➔Help ➔Topics [opens Help Topics Contents window shown in Box 2.14]

➔Find [opens Help Topics Find window shown in Box 2.15] ➔in first box type in the appropriate or closest term [for example, file] ➔Select some matching topics to narrow your search [for example, File] ➔Click a topic, then click Display [for example, Open file] ➔Display [opens Help information box shown in Box 2.16] ➔? [to Minimise or Close the Help system]

If you want help while in a dialog box, select the Help option in the dialog box.
LEAVING SPSS

To leave SPSS, select File and then Exit. If you have edited or analysed data in a session, you will be asked if you wish to save the contents in the data editor or output viewer. If you don’t, select No. If you do, select Yes and name the file if you have not already done so. This means, for example, that if you exit without saving any changes that you may need to keep, those changes will be lost.
Box 2.15 Help Topics Find dialog box

Box 2.16 Help information box


EXERCISES

EXERCISES

1

You need to collect information on the religious affiliation of your respondents. You have thought of the following options: Agnostic, Atheist, Buddhist, Catholic, Jewish, Hindu, Muslim, Protestant and Taoist. Which further category has to be included?

2

You want to record this information in a data file to be stored in a computer. How would you code this information?

3

Looking through your completed questionnaires, you notice that on one of them no answer has been given to this question. What are you going to put in your data file for this person?

4

Suppose that on another questionnaire the respondent had ticked two categories. How would you deal with this situation?

5

The first two of your sample of fifty participants describe themselves as agnostic and the second two as atheists. The ages of these four participants are 25, 47, 33, and 18. How would you arrange this information in your data file?

6

If data were available for all the options of the religious affiliation question, how many columns in an ASCII file would be needed to store this information?

7

How does SPSS know to what the numbers in the data file refer?

8

How many columns to a line are there in most computers for listing data?

9

What is the maximum number of characters that can be used for the name of a variable in SPSS?


APPENDIX 2.1
:
THE JOB SURVEY QUESTIONS
Employee questionnaire

This questionnaire is designed to find out a few things about yourself and your job. Please answer the questions truthfully. There are no right or wrong answers.

Code

Col

1

To which one of the following racial or ethnic groups do you belong? (Tick one)

4

__ White/European

1

__ Asian

2

__ West Indian

3

__ African

4

__ Other

5

2

Are you male or female?

6

Male

1

Female

2

3

What is your current annual income before tax and other deductions?

£________

8–12



4

What was your age last birthday (in years)?

__ years

14–15

5

How many years have you worked for this firm?

__ years

17–18

6

Please indicate whether you (1) strongly disagree, (2) disagree, (3) are undecided, (4) agree, or (5) strongly agree with each of the following statements.

Circle one answer only for each statement.


SD

D

U

A

SA

(a)

I would not leave this firm even if another employer could offer me a little more money

1

2

3

4

5

20

(b)

My job is like a hobby to me

1

2

3

4

5

22

(c)

Most of the time I have to force myself to go to work

1

2

3

4

5

24

(d)

Most days I am enthusiastic about my work

1

2

3

4

5

26

(e)

My job is pretty uninteresting

1

2

3

4

5

28

(f)

I am allowed to do my job as I choose

1

2

3

4

5

30

(g)

I am able to make my own decisions about how I do my job

1

2

3

4

5

32

(h)

People in my section of the firm are left to do their work as they please

1

2

3

4

5

34

(i)

I do not have to consult my supervisor if I want to perform my work slightly differently

1

2

3

4

5

36

(j)

I do my job in much the same way every day

1

2

3

4

5

38

(k)

There is little variety in my work

1

2

3

4

5

40

(l)

My job is repetitious

1

2

3

4

5

42

(m)

Very few aspects of my job change from day to day

1

2

3

4

5

44

7

Did you attend the firm’s meeting this month?

46

__ Yes

1

__ No

2



Supervisor questionnaire

I should be grateful if you would answer the following questions about one of the people for whom you act as supervisor – [Name of employee]

1

Please describe the skill level of work that this person performs. Which one of the following descriptions best fits her/his work? (Tick one)

48


Unskilled

1

__ Semi-skilled

2

__ Fairly skilled

3

__ Highly skilled

4

2

How would you rate her/his productivity? (Tick one)

50

__ Very poor

1

__ Poor

2

__ Average

3

__ Good

4

__ Very good

5

3

How would you rate the quality of her/his work? (Tick one)

52

__ Very poor

1

__ Poor

2

__ Average

3

__ Good

4

__ Very good

5

4

How many days has s/he been absent in the last twelve months?

__ days

54–55

Không có nhận xét nào:

Đăng nhận xét