SMI205 Preregistration form
200382029
2023-05-25
1. My replication project
This is a preregistration of an extension project for the following study:
The paper chosen for the replication project aims to understand the impact of social media and its affect towards trust in the Government in 2012 and 2016 (considering the democratic party occupied the White House for both years). The paper uncovers how polarization politically compare for each party whilst controlling for other factors such as news consumption and demographic
Down below is the reference paper and DOI:
doi number: https://doi.org/10.1080/10584609.2019.1661891
In my replication project I am focusing on the following argument made in the study:
Claim: “We expect a positive realtionship between this variable and our measure of distrust, i.e. increasingly favourable views toward the Republican Party realtive to the Democratic Party should be associated with increasing distrust given that a Democrat occupied the White House in both 2012 and 2016” (page 6)
Summary Statistics representing this:
The moderating role of stealth issue campaigns
2. Planned project extention
2.1. Rationale for new hypothesis
Holt et al. describes the significance of social media affecting political interest based on age. The understanding of social media usage will depend widely on age which in turn can shape people’s political interest. Whilst Holt et al does acknowledge motivations of social media vary between age, they can have a heavy impact and influence on societies political views. The data consisted of Swedish elections (similar to the ANES) and through OLS regression models found younger people shared less interest in politics than older people but found statistically significant results showing social media impacts all ages groups political interest.
2.2. Prediction
Hypothesis: Based on Holt et al’s findings, I will therefore look at how age can impact polarization between both political parties. I expect that the results will include somewhat similar conclusions however, including age as a seperate variable will give this paper a further developed understanding.This can be accomplished by taking age as a seperate vairable and squaring the vairable. From there a prediction model can be produced prepresenting the differences between age. Considering a Democrat occupied the White House in both years I expect that distrust in the government will increase as democrats get older but decrease as republicans get younger
3. Data
In this project I will be using the same data from the original paper (American National Election Studies) including data from the 2012 and 2016 studies
4. Data analysis plan
4.1. Model specification
I will test the hypothesis using the same regression model applied in the study- OLS regression however, aim to encapsulate the differences in age effects between Democrats and Republicans. In the case there is an error with the coding to create two seperate graphs for parties, one single graph will be produced presenting age and social (dis)trust
4.2. Variables
Dependent variable(s):
- Trust in Gov. (trust_index )
Independent variables (IVs):
- party attitudes
- social media usage
- news consumption
- financial condition
- presidential approval
- age as a new separate variable (squared)
- political preference
- stealth inclduing demographic variable
As mentioned in the paper, the dependent variable consists of four items regarding trust on gov. which was then scaled from 0-1.
- A higher score means greater distrust
- A lower score means less distrust
variables have been scaled from 0-1
4.3. Interference criteria
will take the standard acceptance of a p-value greater than 0.05 as being unacceptable and discarded apart from age and age^2 since it is a key variable used to understand the data
4.4. Data exclusion
Outliers will be included. No checks will be performed to determine eligibility for inclusion besides verification that each respondent answered questions
4.5. Missing data
If three of more of the questions are unanswered from the survey, they will be removed from the analysis and data that does not include anything for the dependent variable will be removed
4.6. Exploratory data anlysis
potentially looking for differences in polarization depending on party preference and age
Example: We expect that certain demographic traits may be related to attitudes to migration. Therefore, we will look for relationships between demographic variables (such as age, gender, income) and the attitudinal measures.
5. Session info
This preregistration form was completed in the following R environment:
## R version 4.2.2 (2022-10-31 ucrt)
## Platform: x86_64-w64-mingw32/x64 (64-bit)
## Running under: Windows 10 x64 (build 19045)
##
## Matrix products: default
##
## locale:
## [1] LC_COLLATE=English_United Kingdom.utf8
## [2] LC_CTYPE=English_United Kingdom.utf8
## [3] LC_MONETARY=English_United Kingdom.utf8
## [4] LC_NUMERIC=C
## [5] LC_TIME=English_United Kingdom.utf8
##
## attached base packages:
## [1] stats graphics grDevices datasets utils methods base
##
## loaded via a namespace (and not attached):
## [1] digest_0.6.31 R6_2.5.1 jsonlite_1.8.4 evaluate_0.21
## [5] cachem_1.0.8 rlang_1.1.1 cli_3.6.1 renv_0.17.3
## [9] jquerylib_0.1.4 bslib_0.4.2 rmarkdown_2.21 tools_4.2.2
## [13] xfun_0.39 yaml_2.3.7 fastmap_1.1.1 compiler_4.2.2
## [17] htmltools_0.5.5 knitr_1.42 sass_0.4.6