In this chapter, the results of the regression analysis are presented, disclosing the distribution of intensifiers across time and the groups of speakers within the inferential statistics framework. A negative binomial regression model was applied, which enabled the estimation of the unique contribution of each of four predictors (time, speaker role, gender, and social class) on the frequencies of intensifiers. The chapter explains the rationale behind the choice of the model and presents graphically the results for each predictor as regards the full set of intensifiers as well as within the categories of maximizers, boosters and downtoners. The results show which speaker categories use which kind of intensifiers and which predictors are more important than others for their use.

Chapter 1 tackles ‘foreigners’ defined as such on the basis of their birthplace, as recorded in naval crew musters. Statistical analysis of a sample of 4,392 seamen who served in extra-European stations shows that the proportion of foreign-born men rose between the beginning and the end of the French Wars, likely reflecting the increased demand for manpower. Nearly half of them came not from British imperial or ex-imperial territories, but from continental Europe. Questioning the meaningfulness of categorisations by birthplace, however, the chapter also deploys them as a working hypothesis. The results show that Irish-born, rather than foreign-born, seamen displayed the most distinctive demographic patterns, being on average the oldest but disproportionately employed as ‘landsmen’ – the least skilled and lowest-paid rating. Being born abroad affected the likelihood of promotion to petty officer, and in some cases to able seaman, but it otherwise mattered little in determining a man’s position aboard. Rating was also independent of cultural capital, crudely measured through estimated group numeracy. Overall, a line sharply drawn between the men born in the British Isles and Ireland and those born abroad is a relatively poor predictor of demographic or employment differences.

This topic examines how demand relationships can be estimated from empirical data. The whole process of performing an empirical study is explained, starting from model specification, through the collection of data, statistical analysis and interpretation of results. The focus is on statistical analysis and the application of regression analysis using OLS. Different mathematical forms of the regression model are explained, along with the relevant transformations and interpretations. The concept of goodness of fit, and the coefficient of determination, are explained, along with their application in selecting the best model. The advantages of using multiple regression are discussed, and its implementation and interpretation. Analysis of variance (ANOVA) is explained, and how this relates to goodness of fit. The implications of empirical studies are also discussed, and the light they shed on economic theory. More advanced aspects, related to inferential statistics and hypothesis testing, are covered in an appendix, along with the assumptions involved in the classical linear regression model (CLRM) and consequences of the violation of these assumptions.

In this chapter, there are two types of probabilities that can be estimated: empirical probability and theoretical probability. Empirical probability is calculated by conducting a number of trials and finding the proportion that resulted in each outcome. Theoretical probability is calculated by dividing the number of methods of obtaining an outcome by the total number of possible outcomes. Adding together the probabilities of two different events will produce the probability that either one will occur. Multiplying the probabilities of two events together will produce the probability that both will occur at the same time or in succession. As the number of trials increases, the empirical probability and theoretical probability converge.

It is possible to build a histogram of empirical or theoretical probabilities. As the number of trials increases, the empirical and theoretical probability distributions converge. If an outcome is produced by adding together (or averaging) the results of events, the probability distribution is normally distributed. Because of this, it is possible to make inferences about the population based on sample data – a process called generalization. The mean of sample means converges to the population mean, and the standard deviation of means (the standard error) converges on the value.

This chapter deals with

• The notion of exploratory spatial data analysis

• The presentation of descriptive statistics

• Spatial statistics and their importance in analyzing spatial data

• Analyzing univariate data

• Simple exploratory spatial data analysis tools such as histograms, boxplots and other visual methods to get a better insight of spatial datasets

• Bivariate analysis

• Correlation and pairwise correlation

• Normalization, rescaling and adjustments

• An introduction of basic notions of statistical significant tests

• The importance of hypothesis setting in a spatial context

• The importance of normal distribution in classic statistics and how it is integrated into spatial analysis

After a thorough study of the theory and lab sections, you will be able to

• Have a solid knowledge of descriptive statistics

• Use descriptive statistics for univariate analysis

• Understand and use exploratory spatial data analysis techniques to map and analyze variables attached to spatial objects

• Create several plots, link them to maps and identify interesting patterns of data

• Conduct bivariate analysis, identify whether two variables are linearly related and use plots to further examine their relation

• Rescale data to make comparisons between variables easier and also allow for better data handling

• Apply ESDA tools through ArcGIS and GeoDa

In this chapter, the reader is given a survey of two basic approaches for statistical analysis, the quantitative approach focused on measures of effect size and confidence intervals, and the qualitative approach based on significance values and null hypothesis significance testing.