During the last decade a crisis has emerged in the polling industry because of the rising numbers of missing data. The purpose of this study is to examine the effects of missing data when estimating the public opinion. More specifically, the purpose is to gain understanding of how fluctuations in the amount of missing data affect estimations of political opinion in a Swedish context. Among other methods, we use generalized regression estimation to make our own estimations of the Swedish political opinion. We do this in order to examine the effects of including or not including the variable Voting in the 2018 general election in the estimation model as a way to adjust for lack of response caused by political partisanship. Further, we also examine if certain political events affect the willingness to participate in political opinion polls. Our results show that fluctuations in the amount of missing data does not affect the fluctuations in estimations of political opinion. The reason for this is that the Swedish polling companies already adjust for the lack of response due to political partisanship. We do however find that adjusting for partisanship in estimation models somewhat helps to keep the fluctuations in estimated opinion on an adequate level. Finally, we can also conclude that political events in some cases affect the amount of missing data in political opinion polls. This is true for events that concern all political parties. Our results also show that political events related to a single party or party member does not affect the level of missing data.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:uu-465959 |
| Date | January 2021 |
| Creators | Endredi, David, Lind, Miriam |
| Publisher | Uppsala universitet, Statistiska institutionen |
| Source Sets | DiVA Archive at Upsalla University |
| Language | Swedish |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0027 seconds