Pollster.ge

Last forecasts prior to elections

Between our last forecasts and to the date, several Georgian news outlets published their polls. Following the publication of Formula, Mtavari Arkhi, and Imedi’s polls, we also updated our polling averages. Notably, fieldwork for the latest polls were finished on September 25, meaning that they do not account for events that happened after this date, namely Mikheil Saakashvili’s yet another announcement that he plans returning to Georgia.

According to our forecasts, the Georgian Dream would probably get 37% in elections, following the UNM with 27%, and For Georgia with 9%. Similar to the previous forecast, all other parties poll below 5%.

Looking at temporal trends, by September 25th, GD’s polling numbers were declining, while ratings for the UNM and For Georgia - growing. Nonetheless, presumably events that happened following the publication of the last polls might have affected the way poll numbers move. On top of that, as the ruling party has larger resources of mobilizing voters, a phenomenon that is hard to measure through polls, it is likely that the Georgian Dream might be able to garner more votes than predicted in this forecast.

This might be our last forecast prior to elections, as Friday is the Day of Silence in Georgia. We might update forecasts if covariates in the model (inflation, USD/GEL exchange rate, COVID-19 numbers) shift significantly, or if new polls are publicly available.

Forecasts as of September 5

Forecasts as of September 5 We present the first wave of Pollster.ge forecasts. As of September 5, when the fieldwork for the last poll ended, the ruling Georgian Dream party polled at about 33%, followed by the United National Movement with 25%, For Georgia at 8%, and the Labor party at 6%. Other political parties polled less than 5%. Our forecasts are close to Edson Research’s, as its results are weighted higher in our model.

POST MORTEM: THE 2020 PARLIAMENTARY ELECTIONS, POLLS, AND EXIT POLLS

In the 2020 parliamentary elections, we again faced so-called “polling wars,” yet, opinion research landscape was reasonably diverse. Thirty polls were conducted in the year leading to the 2020 parliamentary elections, with results available to the public. On election day, four different organizations conducted exit polls. How accurate or inaccurate are the pre-election polls and poll-based predictions? In this post, we will review how pre-election polls and exit polls predicted winners.

HYMNS OF REPENTANCE, OR HOW DID OUR FORECASTS WORK?

Prior to the 2021 Municipal Elections of Georgia, Pollster.ge will again offer forecasts based on the publicly available polling data. Thus far, we have tried to predict the outcomes of the 2018 Presidential and 2020 Parliamentary elections. Some of our expectations proved to be accurate, with significant discrepancies. Before we present our forecast for the 2021 elections, we would like to briefly summarize the work we have done. We have applied two different approaches to aggregate polling results, the weighted/running average method and “state space” statistical model.

Forecast for October 28, 2020

Two days before the 2020 Parliamentary Elections of Georgia, Pollster.ge publishes its fifth and final forecasts. Overall, the ruling Georgian Dream party will face challenges in forming the majority in the parliament, while elections most probably will be decided in single-member districts. Our previous forecast showed that ratings for the ruling party were diminishing. The trend was also maintained in the most recent polls too. The weighted averages model indicates that the ruling party’s support shrunk from [47% to 42%](

Forecast for October 14, 2020

The 2020 Parliamentary polls are at the door, and political campaigns are accelerating. So are our forecasts. Our last series of predictions reflected how Georgians felt about political parties in late August. According to them, the Georgian Dream had impressive chances of winning elections and obtaining a comfortable majority in the parliament. Nonetheless, we also warned that the deteriorating epidemiological situation with COVID-19 would force Georgian Dream’s poll numbers to go south.

Forecast for September 18, 2020

Recently, several opinion polls were published in Georgia. Data came from international organizations the National Democratic Institute and the International Republican Institution as well as news media outlets Imedi, Rustavi 2, and Formula. This gave us enough materials to update our forecasts. All surveys that are used in this forecast, were administered prior to the outbreak of Covid-19 in Georgia in early September. We do not have clear understanding how poll numbers were affected, although how volatile Georgian public opinion can get, we do expect some changes.

Forecast for August 17, 2020

Predicted party vote share for August 17th, 2020 This forecast is based on the weighted average method. Data come from 28 publicly available poll results. Proportions are calculated for decided voters only. Poll numbers for four major political parties for August 17th, 2020 The chart below displays poll numbers for four major parties since the late 2016 unti July 24th. Only proportions for decided voters are displayed. State-space model of election forecast for August 17th, 2020 The chart below shows results of 2,000 simulations based on publicly available poll numbers.

Forecast for July 13, 2020

Predicted party vote share for July 13th, 2020 This forecast is based on the weighted average method. Data come from 25 publicly available poll results. Proportions are calculated for decided voters only. Poll numbers for four major political parties for July 13th, 2020 The chart below displays poll numbers for four major parties since late 2016. Only proportions for decided voters are displayed. State-space model of election forecast for July 13th, 2020 The chart below shows results of 2,000 simulations based on publicly available poll numbers.

Polls used in the 2020 parliamentary forecasts

For electoral forecasting, we are currently using thirty-six pre-electoral and political survey results that were conducted after the 2016 parliamentarian elections. Besides, data analysis also includes numbers from the 2016 Parliamentarian and 2017 local government elections. As for the October 2020 wave of CRRC-Georgia’s omnibus survey, we did our version of nonresponse weighting. The corresponding STATA file is available here The range of public opinion surveys is diverse in terms of fieldwork and data collections methodologies, as well as pollsters’ and commissioners’ political and institutional affiliations.

Methodology of forecasts for the 2020 parliamentary elections

Weighted average Our approach is based on two basic methods. The first one, so-called weighted average assumes the averaging of openly available survey results. As you may notice, the data is not comprehensive due to the imperfections of the sources used in our calculations. With rare exceptions, the authors of the surveys seldom publish the methodology, the wording of the questions, and the raw data. The method of forecasting using the weighted average is pretty simple: only the answers indicating the support for a particular political party are taken into account.