Preface page xi
Introduction 1
1 Classifying historical data 7
(a) Nominal data 8
(b) Ordinal data 10
(c) Interval or ratio data 11
(d) Some complications 12
2 Arranging historical data 16
(a) The data set 16
(b) The case 17
(c) The variable 17
(d) The data matrix 18
(e) Collecting data 22
3 Some simple mathematics 27
(a) The frequency distribution 27
(b) Summation notation 33
(c) Logarithms 37
4 The preliminary analysis of data, I: frequency distributions and charts 42
(a) The frequency distribution 43
(b) Cross-classification 48
(c) Charts page 51
(d) Ratio scale graphs 59
5 The preliminary analysis of data, II: summary measures 67
(a) The arithmetic mean 67
(b) The standard deviation 72
(c) The geometric mean 77
(d) The median 77
(e) The mode 81
(f) The coefficient of variation 82
(g) Which to use? 82
6 The analysis of time series 85
(a) Objects and assumptions of the analysis of time series 87
(b) The rate of growth 90
(c) The trend 93
(d) Regular fluctuations in time series 108
(e) The use of ratios and index numbers 117
7 Relationships between variables 125
(a) Is there a relationship? 127
(b) How strong is the relationship? 138
(c) The form of relationships 140
(d) Correlation and regression with time series data 152
8 The problem of imperfect data 155
(a) Too much information: the selection of variables 157
(b) Too much information: the selection of cases 161
(c) The ‘significance’ of sample results 171
(d) Too few data: the problems of missing data 175
(e) Data missing on one or more cases 176
(f) Data missing on one or more variables 178
(g) Data missing on one or more variables of one or more cases, but not on any complete case or variable 182
9 Computers and data processing equipment page 184
(a) Calculating machines 184
(b) Punched card data processing equipment 186
(c) The electronic computer 190
Bibliography 209
Logarithms 212
Antilogarithms 214
Index 216
Biography
Floud, Roderick
