Limits...
The grand old party - a party of values?

Mair P, Rusch T, Hornik K - Springerplus (2014)

Bottom Line: In this article we explore the semantic space spanned by self-reported statements of Republican voters.Our semantic structure analysis uses multidimensional scaling and social network analysis to extract, explore, and visualize word patterns and word associations in response to the stimulus statement "I'm a Republican, because …" which were collected from the official website of the Republican Party.With psychological value theory as our backdrop, we examine the association of specific keywords within and across the statements, compute clusters of statements based on these associations, and explore common word sequences Republican voters use to characterize their political association with the Party.

View Article: PubMed Central - PubMed

Affiliation: Department of Psychology WJH 968, Harvard University, 33 Kirkland St, 02138 Cambridge, MA USA.

ABSTRACT
In this article we explore the semantic space spanned by self-reported statements of Republican voters. Our semantic structure analysis uses multidimensional scaling and social network analysis to extract, explore, and visualize word patterns and word associations in response to the stimulus statement "I'm a Republican, because …" which were collected from the official website of the Republican Party. With psychological value theory as our backdrop, we examine the association of specific keywords within and across the statements, compute clusters of statements based on these associations, and explore common word sequences Republican voters use to characterize their political association with the Party.

No MeSH data available.


MDS solution based on constant input dissimilarities.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4256162&req=5

Fig1: MDS solution based on constant input dissimilarities.

Mentions: As mentioned, in our application on political values we operate on a sparse DTM. Note that such a sparse DTM setting is very common for text data and, therefore, makes our approach applicable to a wide variety of sparse and/or DTM-based MDS applications. The problem with such data is that using cosine distances (or other well-known distance measures such as Euclidean distance or Jaccard distance) lead to a so called "special MDS" solution since the dissimilarities have a very low variance: i.e. they are "almost equal". (Borg and Groenen2005) call it, "a special solution because of almost equal dissimilarities" (see also Buja et al.1994). Applying an MDS based on standard distance measures would lead to a concentric, circular representation of the configuration in the low-dimensional space that does not reveal or reproduce structures in the observed data matrix. Figure1 shows such a solution based on constant input dissimilarities. Any permutation of the labels over the points give another local minimum, so that the labels can be almost arbitrarily be assigned to the points. We solve this problem by the using a word co-occurrence based gravity approach.Figure 1


The grand old party - a party of values?

Mair P, Rusch T, Hornik K - Springerplus (2014)

MDS solution based on constant input dissimilarities.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4256162&req=5

Fig1: MDS solution based on constant input dissimilarities.
Mentions: As mentioned, in our application on political values we operate on a sparse DTM. Note that such a sparse DTM setting is very common for text data and, therefore, makes our approach applicable to a wide variety of sparse and/or DTM-based MDS applications. The problem with such data is that using cosine distances (or other well-known distance measures such as Euclidean distance or Jaccard distance) lead to a so called "special MDS" solution since the dissimilarities have a very low variance: i.e. they are "almost equal". (Borg and Groenen2005) call it, "a special solution because of almost equal dissimilarities" (see also Buja et al.1994). Applying an MDS based on standard distance measures would lead to a concentric, circular representation of the configuration in the low-dimensional space that does not reveal or reproduce structures in the observed data matrix. Figure1 shows such a solution based on constant input dissimilarities. Any permutation of the labels over the points give another local minimum, so that the labels can be almost arbitrarily be assigned to the points. We solve this problem by the using a word co-occurrence based gravity approach.Figure 1

Bottom Line: In this article we explore the semantic space spanned by self-reported statements of Republican voters.Our semantic structure analysis uses multidimensional scaling and social network analysis to extract, explore, and visualize word patterns and word associations in response to the stimulus statement "I'm a Republican, because …" which were collected from the official website of the Republican Party.With psychological value theory as our backdrop, we examine the association of specific keywords within and across the statements, compute clusters of statements based on these associations, and explore common word sequences Republican voters use to characterize their political association with the Party.

View Article: PubMed Central - PubMed

Affiliation: Department of Psychology WJH 968, Harvard University, 33 Kirkland St, 02138 Cambridge, MA USA.

ABSTRACT
In this article we explore the semantic space spanned by self-reported statements of Republican voters. Our semantic structure analysis uses multidimensional scaling and social network analysis to extract, explore, and visualize word patterns and word associations in response to the stimulus statement "I'm a Republican, because …" which were collected from the official website of the Republican Party. With psychological value theory as our backdrop, we examine the association of specific keywords within and across the statements, compute clusters of statements based on these associations, and explore common word sequences Republican voters use to characterize their political association with the Party.

No MeSH data available.