Sammenlign metoder
Gjennomgå de valgte metodene side om side; rader som avviker, er uthevet.
| Multidimensional Unfolding× | NOMINATE× | |
|---|---|---|
| Fagfelt | Political Science | Political Science |
| Familie | Latent structure | Latent structure |
| Opprinnelsesår≠ | 2000 | 1985 |
| Opphavsperson≠ | Keith T. Poole (nonparametric optimal classification and unfolding) | Keith T. Poole and Howard Rosenthal |
| Type≠ | Latent-space scaling model placing individuals and stimuli in a joint space | Spatial scaling model of roll-call voting |
| Opprinnelig kilde≠ | Poole, K. T. (2000). Nonparametric Unfolding of Binary Choice Data. Political Analysis, 8(3), 211–237. DOI ↗ | Poole, K. T., & Rosenthal, H. (1985). A Spatial Model for Legislative Roll Call Analysis. American Journal of Political Science, 29(2), 357–384. DOI ↗ |
| Alias | Unfolding analysis, Optimal classification, Preference unfolding, Joint-space scaling | DW-NOMINATE, W-NOMINATE, Nominal Three-Step Estimation, Poole-Rosenthal scores |
| Relaterte≠ | 5 | 3 |
| Sammendrag≠ | Multidimensional unfolding places both individuals and the stimuli they evaluate — candidates, parties, bills — in a single joint low-dimensional space, so that each person's preferences are explained by their proximity to the stimuli. In political science it underlies Keith Poole's nonparametric optimal classification of roll-call votes and the unfolding of thermometer ratings and rank orders, recovering legislators' and bills' positions from nothing but the pattern of choices. Unlike correlation-based scaling, unfolding treats preference as a single-peaked function of distance: you like what is close to you and dislike what is far. | NOMINATE — Nominal Three-step Estimation — is the family of spatial scaling procedures developed by Keith Poole and Howard Rosenthal to recover legislators' ideological positions from roll-call votes. Each legislator and the yea and nay outcomes of each vote are placed in a low-dimensional space, and a normal (Gaussian) deterministic utility plus a random shock governs choices. Fitted by maximum likelihood, NOMINATE produces the canonical ideal-point coordinates used to chart polarization across two centuries of the U.S. Congress, with the dynamic DW-NOMINATE variant allowing positions to drift smoothly over time. |
| ScholarGateDatasett ↗ |
|
|