single row, given the portrait orientation of journals. The
x-axis depicts which model is being displayed. To facilitate
comparison across predictors, we center the y-axis at zero,
which is the null hypothesis for each of the predictors.
The regression table presents six models, which vary
with respect to sample (full sample vs. excluding partisans
registration counties) and predictors (with/without state
year dummies and with/without law change). On the x-axis
we group each type of model: “full sample,” “excluding
counties with partial registration” and “full sample with
state year dummies.” Within each of these types, two dif-
ferent regressions are presented: including the dummy vari-
able law change and not including it. Thus, for each type,
we plot two point estimates and inter vals—we differenti-
ate the two models by using solid circles for the models in
which law change is included and empty circles for the
models in which it is not. We again choose not to graph
the estimates for the constants because they are not sub-
stantively meaningful.
This graphing strategy allows us to easily compare point
estimates and confidence intervals across models. Although
in all the specified models the percent of county with regis-
tration predictor is statistically significant at the 95 per-
cent level, it is clear from the graph that estimates from
the full sample with state/year dummies models are sig-
nificantly different from the other four models. In addi-
tion, by putting zero at the center of the graph, it becomes
obvious which estimates have opposite signs depend-
ing on the specification (log population and log median
family income). By contrast, it is much more difficult to
spot these changes in signs in the original table. Thus, by
using a graph it is easy to visually assess the robustness of
each predictor—both in terms of its magnitude and con-
fidence interval—simply by scanning across each panel.
In summary, the graph appropriately highlights the insta-
bility in the estimates depending on the choice of model.
Table 8
Pekkanen, Nyblade and Krauss (2006),
table 1: Logit analysis of electoral
incentives and LDP post allocation
(1996–2003)
Variable Model 1 Model 2
Block 1: MP Type
Zombie 0.18 (.22) 0.27 (0.22)
SMD Only −0.19 (0.22) −0.19 (0.24)
PR Only −0.39 (0.18)** —
Costa Rican in PR −0.09 (0.29) —
Block 2: Electoral Strength
Vote share margin — 0.005 (0.004)
Margin Squared — —
Block 3: Misc Controls
Urban-Rural Index 0.04 (0.08) 0.04 (0.09)
No Factional
Membership
−0.86 (0.26)*** −0.98 (0.31)***
Legal Professional 0.39 (0.29) −.36 (0.30)
Seniority
1
st
Term −3.76 (0.36)*** −3.66 (0.37)***
2
nd
Term −1.61 (0.19)*** −1.59 (0.21)***
4
th
Term −0.34 (0.19)** −0.45 (0.21)***
5
th
Term −1.17 (0.22)*** −1.24 (0.24)***
6
th
Term −1.15 (0.22)*** −1.04 (0.24)***
7
th
Term −1.52 (0.25)*** −1.83 (0.29)***
8
th
Term −1.66 (0.28)*** −1.82 (0.32)***
9
th
Term −1.34 (0.32)*** −1.21 (0.33)***
10
th
Term −2.89 (0.48)*** −2.77 (0.49)***
11
th
Term −1.88 (0.43)*** −1.34 (0.46)***
12
th
Term −1.08 (0.41)*** −0.94 (0.49)**
Constant .020 (.20) 0.13 (0.26)
Log-likelihood −917.24 −764.77
N 1895 1574
Notes: Dependent Variables: 1 if MP holds a post of minister,
vice minister, PARC, or HoR Committee Chair.
Base categories: SMD dual-listed, 3rd term. Excluded obser-
vations: senior MPs that held no post (> 12 terms, PR-Only
MPs in Model 2).
*p < .10, **p < .05, ***p < .001.
Figure 7
Using parallel dot plots with error bars to
present two regression models.
Table 1 from Pekkanen et al. 2006 displays two logistic regres-
sion models that examine the allocation of posts in the LDP party
in Japan. We turn the table into a graph, and present the two mod-
els by plotting parallel lines for each of them grouped by coef-
ficients. We differentiate the models by plotting different symbols
for the point estimates: filled (black) circles for Model 1 and
empty (white) circles for Model 2.
| |
!
!
!
December 2007
|
Vol. 5/No. 4 767