Inledning Plus/minus Studier
Att arbeta med registerdata
Pär Nyman Statsvetenskapliga institutionen
Forskargruppen i hälso-
och sjukvårdsforskning
20 September, BMC
Inledning Plus/minus Studier
Inledning
Bakgrund
CONPOL-projektet (conpol.org).
Hur vi arbetar med registerdata.
• För- och nackdelar.
• Typer av studier.
• Anpassar gärna.
Vad är registerdata? Typexempel:
• Förs av myndigheter.
• Innehåller hårda/objektiva data.
• Över jättemånga individer.
- 2 -
Inledning Plus/minus Studier
Våra data
Projektets server har data på Demografi (inkl. bostadsområde).
Socioekonomiska variabler (inkl. skola och arbetsplats).
Mönstringsdata (kognitiv och social förmåga).
Politiskt deltagande (valdeltagande och politiker).
Andra populära typer av data Belastningsregistret.
Hälsodata.
Inledning Plus/minus Studier
För- och nackdelar
Fördelar
Antalet observationer.
• Precisa skattningar.
I Obegränsat antal variabler.
I Heterogenitetsanalyser.
• Inget bortfall.
• Populationsdata.
Objektiva och precisa mått.
Fortfarande en säljande faktor (och motiverar Sverige).
Större möjligheter till kausala analyser (men inte så enkelt).
- 4 -
Inledning Plus/minus Studier
För- och nackdelar
Nackdelar
Dålig dokumentation (men läs den som finns!).
Sällan mått på det vi vill ha.
• Begränsningar.
• Validitetsproblem.
• Kreativitetsdödande? Bekvämt?
Omständligt.
• Dyrt och tidskrävande att införskaffa.
• Tar tid att förbereda och analysera data.
• Delad infrastruktur ställer större krav på samarbete.
• Inte bara positivt att man kan göra hundra andra analyser.
Inledning Plus/minus Studier
Studier
Typiska studier
Deskriptiva/kartläggande studier.
Kausala studier (fr.a. naturliga experiment).
Liten typologi
Kontrollerade experiment (randomisering).
Naturliga experiment (så gott som randomiserat).
• Diskontinuiteter (RD).
• Reformer.
- 6 -
Inledning Plus/minus Studier
Studier
Typiska studier
Deskriptiva/kartläggande studier.
Kausala studier (fr.a. naturliga experiment).
Liten typologi
Kontrollerade experiment (randomisering).
Naturliga experiment (så gott som randomiserat).
• Diskontinuiteter (RD).
• Reformer.
Inledning Plus/minus Studier
Studier
Dancygier (2015) Why are immigrants underrepresented in politics?
Andelen utrikes födda bland politiker är inte ens hälften så hög som deras andel av befolkningen.
Två huvudförklaringar:
• Individers resurser.
• Politiska kontexten.
Tidigare forskning har dock saknat data om politikerna.
- 7 -
Inledning Plus/minus Studier
Studier
Dancygier (2015) Why are immigrants underrepresented in politics?
Why Are Immigrants Underrepresented in Politics? November 2015FIGURE 6. Determinants of the Native-Immigrant Representation Gap; Results from the Oaxaca-Blinder Decomposition
-.05 0
.05.1.15.2
1991 1994 1998 2002 2006 2010
Unexplained Seats to Voters Opportunity Structure SES Demographics
about 22 percent of the overall representation gap is attributable to this variable. In 2010 this figure reached 36 percent.
Figure 6
displays these patterns and summarizes the findings. The sources of the representation gap have shifted over the years, from differences in returns to basic socio-demographic factors at the beginning (the unexplained portion typically attributed to discrimi- nation), towards a situation where group differences in the distribution of these characteristics are increasingly influential. Specifically, the fact that immigrants, more so than natives, live in large municipalities with low seats-to-voter ratios has become more important.
In line with these results, we also find that a sig- nificant and negative immigrant effect remains when we match immigrants and natives on all the variables included in
Table 1(except for Citizenship and Time
in Country), but that this immigrant penalty declinessubstantially over time. With the matching technique, the magnitude of the immigrant effect decreases by almost two-thirds between 1991 and 2010, which cor- responds fairly well with the drop in the unexplained part found when using the OB decomposition (see the Online Appendix).
29IS DISCRIMINATION AN IMPORTANT DRIVER OF THE REPRESENTATION GAP?
In this section we probe whether it is plausible to as- sume that discrimination by party gatekeepers is at
29For reference, the Online Appendix also includes results that were obtained by only matching on those covariates (gender and age) that are not affected by immigrant status. The general pattern is similar to that found in the main results, with a substantial decrease in the immigrant effect over time.
least partly responsible for the sizable unexplained portion of the representation gap. Our main strategy will be to turn from the cross-sectional evidence that we have thus far relied on to movements in the unex- plained portion of the representation gap over time.
Stated differently, we view the unexplained part as an estimate of the upper bound of discrimination and now investigate whether discrimination is a plausible expla- nation for why and how this estimate changes over time.
To do so, we assess five mechanisms: First, we ex- amine a competing hypothesis, namely that increased political interest and participation among immigrants leads to a narrowing of the representation gap. Sec- ond, we discuss the possibility that the decrease in the representation gap is due to trends in the electoral suc- cess of parties that are more inclusive of immigrants.
Third, we turn to the behavior of natives and provide three pieces of evidence that are consistent with the notion that discrimination is critical but also declining over time. First, we show that native attitudes towards immigrants are relatively hostile, but that they have be- come more favorable over time. Next, we demonstrate that immigrants who encounter more discrimination in other realms—those from less developed and cul- turally more distant countries—are also less likely to win office, but that their chances have improved. Last, we show that party gatekeepers are prone to placing immigrants on unfavorable list positions. Yet, this bias diminishes over the years, making it more likely that nominated immigrant candidates actually win office.
Though we conceptualize discrimination broadly as equal cases being treated unequally in the elec- toral process, the unexplained part of the OB de- composition captures all unobserved heterogeneity, not just variation caused by unequal returns to the socio-demographic and human capital variables we
Decomposing the under-representation
- 7 -
Inledning Plus/minus Studier
Studier
Dancygier (2015) Why are immigrants underrepresented in politics?
Fin publicering bl.a. tack vare populationsdata på röstberättigade och politiker.
Kan dock inte utesluta skillnader som beror på politiskt intresse eller självtillit.
Vi arbetar nu med en kompletterande studie baserad på enkätdata.
- 7 -
Inledning Plus/minus Studier
Studier
Aggeborn och Nyman (2019) Intergenerational transmission of party affiliation
Vanligt att toppolitiker kommer från en politisk familj.
Har de gjort ett självständigt val av parti?
Knyter an till en gammal litteratur om politisk socialisering.
Resultat: 76 procent kandiderar för samma parti som sina föräldrar.
Huvudproblem: Kan inte separera ideologi från selektion (eftersom politiker är ett selekterat urval).
Verkar i alla fall drivas av uppväxtfaktorer.
Inledning Plus/minus Studier
Studier
Folke och Ricke (2017) All the single ladies Ekonomisk jämställdhet sämst på toppjobb.
H: Svårare för kvinnor än män att kombinera privatliv med krävande arbete.
RQ: Ökar risken för skilsmässa när kvinnor blir befordrade?
Korrelation är inte kausalitet och svårt att randomisera toppjobb.
- 9 -
Inledning Plus/minus Studier
Studier
Folke och Ricke (2017) All the single ladies KSO-kandidater (listettorna för största partierna).
Tydlig diskontinuitet.
Vid jämna val visar balanstest inga skillnader före valet.
Inledning Plus/minus Studier
Studier
Folke och Ricke (2017) All the single ladies
18 lines in the figure represent the share of persons who remain married to their spouse in two groups:
those who are promoted (black lines) and those who are not promoted (gray lines). Both lines start at the value "1" at the left-hand side of the graphs at t = -3, three years prior to the election. The value 1 indicates, of course, that our sample consists of 100% married individuals (also recall that the election takes place in time period t = 0 and the person's first year in office is t = 1).
Figure 7. Marriage durability and divorce among men and women
Over time, all lines slope downward as some proportion of marriages falls apart each year.
Before the promotion event, which is marked in the figure by a vertical black line, the rate of divorce is highly similar between the winners and losers of the (subsequent) promotion contest. The main result is the fact that in the first three years after the promotion, the black and gray lines diverge in the plots for the sub-sample of women, but not in the sub-sample of men. Among women, the rate of divorce increases among the promoted women compared to those who were not promoted to create a difference of about 7 to 8 percentage points three years after the promotion event. These groups do not reconverge later. The absence of different divorce rates between the (subsequently) promoted and unpromoted women before the promotion event strongly suggests that this is a causal effect.
Furthermore, the pre-trends are highly similar in the full sample of elections, which suggests that later analysis that relies on the full sample does not suffer from serious endogeneity problems.