A comparison of statistical methods for the aggregation of regional linguistic variation

Grieve, Jack

doi:10.1515/9783110317558.53

Cited by 37 publications

(10 citation statements)

References 21 publications

(42 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Once the syntactic distances have been computed for all survey site pairs in the dataset, it is then possible to compute the correlation between the linguistic and the geographic distances over all survey site pairs. We use three correlation measures: two measures of linear correlation—the Pearson product-moment correlation coefficient and the distance-oriented Mantel test (Mantel, 1967), following Scherrer (2012), Haynie (2012), and Grieve (2014)— and logarithmic correlation. Furthermore, regression models of different types (linear, logarithmic) are fitted to the distributions of syntactic distance against geographic distance, separately for each type of geographic distance measure used.…”

Section: Methodsmentioning

confidence: 99%

“…We use three correlation measures: two measures of linear correlationthe Pearson product-moment correlation coefficient and the distance-oriented Mantel test (Mantel, 1967), similarly to Scherrer (2012), Haynie (2012) and Grieve (2014)-and logarithmic correlation. Furthermore, regression models of different types (linear, logarithmic) are fitted to the distributions of syntactic distance against geographic distance, separately for each type of geographic distance measure used.…”

Section: Correlation Of Syntactic and Geographic Distancesmentioning

confidence: 99%

See 1 more Smart Citation

Exploring global and local patterns in the correlation of geographic distances and morphosyntactic variation in Swiss German

Jeszenszky

Stoeckle

Glaser

et al. 2017

J. of Ling. Geography

View full text Add to dashboard Cite

Using data from a Swiss German dialect syntax survey, this study aims to explore, in a spatially differentiated manner, the correlation between dialectal variation and geographic distances. A linguistic distance was expressed by a measure aggregated from 60 survey questions. To operationalize the possibility of language contact, Euclidean distance, as well as travel times in 2000, 1950 and 1850 between survey sites were used. Going beyond previous work by others, we also explore the covariation of geographic and linguistic distances at the local level, focusing on spatial subsets and individual survey sites, thus being able to paint a more differentiated picture. With the diverse physical landscape of Switzerland making an impact on potential language contact, we find that travel times are a better predictor than Euclidean distance for the syntactic variation in Swiss German dialects. However, on the local scale the difference is not always significant, depending on prevalent topography. AbstractUsing data of a Swiss German dialect syntax survey this study aims at exploring, in a spatially differentiated manner, the correlation between dialectal variation and geographic distances. A linguistic distance was expressed by a measure aggregated from 60 survey questions. To operationalise the possibility of language contact, Euclidean distance and travel times in 2000, 1950 and 1850 between survey sites were used. Going beyond previous work by others, we also explore the covariation of geographic and linguistic distances at the local scale, focusing on spatial subsets and individual survey sites, thus being able to paint a more differentiated picture. With the diverse physical landscape of Switzerland making an impact on potential language contact, we find that travel times are a better predictor than Euclidean distance for the syntactic variation in Swiss German dialects. However, on the local scale the difference is not always significant, depending on prevalent topography.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Correlation Of Syntactic and Geographic Distancesmentioning

confidence: 99%

Exploring global and local patterns in the correlation of geographic distances and morphosyntactic variation in Swiss German

Jeszenszky

Stoeckle

Glaser

et al. 2017

J. of Ling. Geography

View full text Add to dashboard Cite

show abstract

“…In contrast to measures of global spatial autocorrelation, which return a single-value z score for each variable (indicating the degree of regional clustering across the entire distribution of that variable), measures of local spatial autocorrelation produce one value for each location for each variable, indicating the degree to which that particular location is part of a hot spot or cold spot [ 37 ].…”

Section: Methodsmentioning

confidence: 99%

Spatial Distribution of Underweight, Overweight and Obesity among Women and Children: Results from the 2011 Uganda Demographic and Health Survey

Turi

Christoph

Grigsby-Toussaint

2013

IJERPH

View full text Add to dashboard Cite

While undernutrition and infectious diseases are still persistent in developing countries, overweight, obesity, and associated comorbidities have become more prevalent. Uganda, a developing sub-Saharan African country, is currently experiencing the public health paradox of undernutrition and overnutrition. We utilized the 2011 Uganda Demographic and Health Survey (DHS) to examine risk factors and hot spots for underweight, overweight, and obesity among adult females (N = 2,420) and their children (N = 1,099) using ordinary least squares and multinomial logit regression and the ArcGIS Getis-Ord Gi* statistic. Overweight and obese women were significantly more likely to have overweight children, and overweight was correlated with being in the highest wealth class (OR = 2.94, 95% CI = 1.99–4.35), and residing in an urban (OR = 1.76, 95% CI = 1.34–2.29) but not a conflict prone (OR = 0.48, 95% CI = 0.29–0.78) area. Underweight clustered significantly in the Northern and Northeastern regions, while overweight females and children clustered in the Southeast. We demonstrate that the DHS can be used to assess geographic clustering and burden of disease, thereby allowing for targeted programs and policies. Further, we pinpoint specific regions and population groups in Uganda for targeted preventive measures and treatment to reduce the burden of overweight and chronic diseases in Uganda.

show abstract

“…In parallel, novel visualization and data analysis techniques from spatial statistics were introduced, such as hierarchical clustering (Goebl 1984), multi-dimensional scaling (Embleton 1993), and correlation analysis (Heeringa & Nerbonne 2001;Goebl 2005). For a comparative review of methods used in dialectometry, see Grieve (2014) and Wieling & Nerbonne (2015). Kelle (2001) provides the first dialectometric account of German-speaking Switzerland.…”

mentioning

confidence: 99%

A quantitative approach to Swiss German – Dialectometric analyses and comparisons of linguistic levels

Scherrer

Stoeckle

2016

Dialectologia Et Geolinguistica

View full text Add to dashboard Cite

German-speaking Switzerland can certainly be regarded as one of the liveliest and at the same time best researched dialect areas in Central Europe. It is all the more surprising that dialectometric analyses in this area have only recently been performed and none of them included an investigation into the level of syntax. In this paper we pursue two goals: First, we present digital data that has been made available in recent years on the basis of the Sprachatlas der deutschen Schweiz (SDS) and the Syntaktischer Atlas der deutschen Schweiz (SADS). Our second goal is to present dialectometric analyses performed with this data. A special focus is put on the comparison of different linguistic levels (lexis, phonology, morphology and syntax). Our methods include hierarchical cluster analyses (of the whole dataset as well as of the linguistic levels), correlations (between pairs of linguistic levels and between linguistic levels and geography) and parameter maps which allow us to draw conclusions about the distributions of innovative and conservative regions, dialect centers and transition zones. Our results show that while all four levels generally yield similar geographic patterns (dynamic areas in the North vs. conservative areas in the South, agreement of dialect and cantonal borders, high correlations with geography), syntax deviates most from the other levels.

show abstract

A comparison of statistical methods for the aggregation of regional linguistic variation

Cited by 37 publications

References 21 publications

Exploring global and local patterns in the correlation of geographic distances and morphosyntactic variation in Swiss German

Exploring global and local patterns in the correlation of geographic distances and morphosyntactic variation in Swiss German

Spatial Distribution of Underweight, Overweight and Obesity among Women and Children: Results from the 2011 Uganda Demographic and Health Survey

A quantitative approach to Swiss German – Dialectometric analyses and comparisons of linguistic levels

Contact Info

Product

Resources

About