Doing Research in Psychology (PSYCHOL 2004)

1. Aims and Hypothesis

The main aim of this project is to analyze the provided social media dataset with the help of R programming. Currently social media has a vital role in exposing and spreading wrong information to people across various domains such as science, diet, politics, health and so on. There are various levels of supporting approaches to identify misinformation such as training, censorship, warnings, face checking etc. Based on this, the hypotheses will be formulated to investigate the social media data with countering misinformation. The formulated hypotheses are listed:

Hypothesis One: Two Groups with One Predicting Outcome

• H0: Facebook, and Instagram Misinformation Warning Support Mental health is comparatively low.

• H1: Facebook, and Instagram Misinformation Warning Support Mental health is comparatively high.

According to the above hypothesis, The selected two groups are Facebook and Instagram, predicted outcome variable as Misinformation Warning Support Mental health.

Hypothesis Two: Three Groups with One Predicting Outcome

• H0: Facebook, Instagram, and YouTube Misinformation removal support US politics is comparatively low.

• H1: Facebook, Instagram, and YouTube Misinformation removal support US politics is comparatively high.

According to the above hypothesis, the selected three groups are Facebook, YouTube, and Instagram i.e., independent variables, predicted outcome i.e., dependent variable as misinformation removal support US politics.

Hypothesis Three: Difference between two numeric variables

• H0: There is no difference between Facebook and YouTube Misinformation of About Vaccine.

• H1: There is a difference between Facebook and YouTube Misinformation of About Vaccine.
  According to the above hypothesis, the selected two numeric variables are Facebook misinformation vaccine, and YouTube misinformation vaccines. To address this, T-test statistical analysis is used (GeeksforGeeks, 2023).

Hypothesis Four: Relationship between two numeric variables

• H0: There is no relationship between open minded thinking and general conspiracist thinking.

• H1: There is a relationship between open minded thinking and general conspiracist thinking.

From the above hypothesis, the selected two numerical variables are open minded thinking and general conspiracist thinking. Based on the two numerical variables, the linear regression will be performed to determine the relationship between the two numerical variables.

2. Methods

2.1Participants

A research about misinformation spreading via social media is considered, which is analyzed using machine learning methods for conducting analysis. To help this study, a survey is conducted, where there were 180 participants. And, this survey had no pre-prescribed questions. The data provides a range of variables associated to the role of how social media can impact in spreading and exposing people by using misinformation. The data is expected to determine the habits of social media and exposure to misinformation throughout different domains.

The research findings are expected to reveal if misinformation spreading is influenced by social medias content or not. If it is true, which is the respective platform that influences this factor. Based on this determination, actions can be taken to control the spread of misinformation.

This survey aims to investigate the social medias role is spreading misinformation. Based on the provided variables the hypotheses are formulated. There are eleven variables such as follows:

a) id

b) Age

c) Gender

d) Platform_use_***

e) Platform_frequency_***

f) Platform_time_***

g) ***_Misinfo_###

h) Labelling_Condition

i) ^^^_Support_###

j) OpenMindedThinking

k) ConspiracistThinking

Statistical test is conducted to determine the results. The results are concluded on the basis of models p-value and its statistical significance.

2.2Materials

Today, it is a fact that social networks contain a lot of misinformation, and it can be misleading to the people, and they end up taking wrong decisions, or build negative options and causes social threats.

A survey was conducted, which had 180 participants. This research includes assessment of data collected by students in a survey and shared in the class. The student has constructed suitable aims and hypotheses based on the provided data, selected suitable methods for conducting the analysis and determine the results to conclude the outcome of the data analysis.

2.3Procedure

In this report, hypothesis testing is considered to answer the formulated hypothesis with the help of R programming.

The following models are created and tested using hypothesis testing:

1) Multiple linear regression model

2) Simple linear regression model

3) T-test

The aim of this survey is to construct a series of aims and hypotheses depending the data using particular statistical tests. For this, four hypotheses will be created. Then, relevant statistical analyses need to be run for testing the created hypotheses. The findings are determined from the values obtained from the p-value of the created model. Further, the statistical significance of the model is determined, where if the value is lower than 0.05 it denotes having a statistically significant model, whereas, if the value is higher than 0.05 it denotes the model is not statistically significant. Based on the p-value, the created hypotheses reject the null hypothesis and accept the alternative hypothesis accordingly.

In the end, based on this determination, the recommendations to prevent the spread of misinformation is needed or not can be detected and help to stop the spread of misinformation on social media platform with effective measures.

3. Results

The hypothesis testing will be performed to answer the formulated hypothesis with the help of R programming.

To begin with, address the first hypothesis:

• H0: Facebook, and Instagram Misinformation Warning Support Mental health is comparatively low.

• H1: Facebook, and Instagram Misinformation Warning Support Mental health is comparatively high.

To address this, multiple linear regression model is used by selecting the dependent variable as warning support mental health, and independent variables are platform use Facebook and platform use Instagram. The result of multiple linear regression is demonstrated below (Keita, 2022).

The result in the above figure shows that the p-value of linear regression model is 0.00 and it is lower than 0.05, and it means to have a statistically significant model. Here, 0.01 is the value of R square and it is a low percentage of variance for the dependent variable, which describes independent variables. This helps in measuring the goodness of fit and relationship between the created model and the dependent variable on a convenient scale (ranging from 0 to 100%). Here, 0.01 is the r squared value, and it denotes that 1% of the dependent variable explaining the independent variable, which denotes to have an unfit model for this prediction. Based on the p-value, reject null hypothesis and accept the alternative hypothesis. Therefore, it concludes that Facebook, and Instagram Misinformation Warning Support Mental health is comparatively high. It means that Facebook and Instagram highly spread misinformation about mental health.

Moving on, addresses the second hypothesis:

• H0: Facebook, Instagram, and YouTube Misinformation removal support US politics is comparatively low.

• H1: Facebook, Instagram, and YouTube Misinformation removal support US politics is comparatively high.

To address this, multiple linear regression model is used by selecting the dependent variable as removal support US politics, and independent variables are platform use Facebook, platform use YouTube and platform use Instagram. The result of multiple linear regression is demonstrated below.

According to the above result, the created linear regression model has 0.00 of p-value, which is less than 0.05, indicates that the created model is statistically significant. 0.02 is the value of R square and it is a low percentage of variance for the dependent variable, which describes independent variables. It denotes the dependent variable has 2% explaining the independent variable and means to have an unfit model for this prediction. Based on p-value, rejection of null hypothesis and acceptance of alternative hypothesis. Therefore, it concludes that Facebook, Instagram, and YouTube Misinformation removal supporting US politics is comparatively high. In other words, it means Facebook, YouTube and Instagram highly spread misinformation about the US politics regarding Warning supports.

Next, address the following third hypothesis:

• H0: There is no difference between Facebook and YouTube Misinformation of About Vaccine.

• H1: There is a difference between Facebook and YouTube Misinformation of About Vaccine.

Even to address this, T-test is used, as it will determine the difference between two means. In our case, two variables are selected such as, Facebook misinformation vaccine and YouTube misinformation vaccines. The result of T-test is demonstrated below (Soetewey, 2020).

As per the above result, the t statistics is 2.991, degree of freedom is 243.26, and p-value is 0.03, which indicates that strong evidence to null hypothesis is false and alternative hypothesis is true. And, it determined that average of Facebook misinformation about vaccine is 2.60, and average of YouTube misinformation about vaccine is 2.01, it means that participant strongly disagree that there is no spreading of misinformation about vaccine on Facebook and YouTube. In contrast to Facebook and YouTube, most of the participants agreed that YouTube platform does not spread misinformation about vaccine when compared with Facebook. So far, it shows that there is a difference between YouTube and Facebook misinformation about vaccine. Further, based on the p-value, we have to reject null hypothesis and accept the null hypothesis. Therefore, it concludes that there is a difference between Facebook and YouTube Misinformation of About Vaccine.

At last, address the fourth hypothesis:

• H0: There is no relationship between open minded thinking and general conspiracist thinking.

• H1: There is a relationship between open minded thinking and general conspiracist thinking.

To address this hypothesis, simple linear regression model is used by selecting the dependent variable as open-minded thinking and independent variable as general conspiracist thinking. The result of simple linear regression model is demonstrated below.

The obtained result shows that 0.00 is the p-value of the created linear regression model, and it is lesser than 0.05. It means to have a statistically significant model. The, 0.03 is the value of R square and it is a low percentage of variance for the dependent variable, which describes independent variables. It signifies to have 3% of the dependent variable explaining the independent variable, which denotes to have an unfit model for this prediction. Based on p-value, reject null hypothesis and accept the alternative hypothesis. Therefore, it concludes that there is a relationship between open minded thinking and general conspiracist thinking.

4. Conclusion

In this report, four hypotheses are framed and their results are described as follows:

First Hypothesis:

Here, multiple linear regression model was used

H0: Facebook, and Instagram Misinformation Warning Support Mental health is comparatively low.

H1: Facebook, and Instagram Misinformation Warning Support Mental health is comparatively high.

The result concluded that Facebook, and Instagram Misinformation Warning Support Mental health is comparatively high, which showed that Facebook and Instagram highly spread misinformation about mental health.

Second Hypothesis:

H0: Facebook, Instagram, and YouTube Misinformation removal support US politics is comparatively low.

H1: Facebook, Instagram, and YouTube Misinformation removal support US politics is comparatively high.

Here, multiple linear regression model was used. This result concluded that Facebook, Instagram, and YouTube Misinformation removal supporting US politics is comparatively high. It indicates Facebook, YouTube and Instagram highly spread misinformation about the US politics regarding Warning supports.

Third Hypothesis:

H0: There is no difference between Facebook and YouTube Misinformation of About Vaccine.

H1: There is a difference between Facebook and YouTube Misinformation of About Vaccine.

T-test was used, and the result concluded that there is a difference between Facebook and YouTube Misinformation of About Vaccine.

Fourth Hypothesis:

H0: There is no relationship between open minded thinking and general conspiracist thinking.

H1: There is a relationship between open minded thinking and general conspiracist thinking.

Here, simple linear regression model was used. The result concluded that there is a relationship between open minded thinking and general conspiracist thinking.

Thus, the result of this report has concluded having a statistically significant relationship between open minded thinking and general conspiracist thinking, which is determined using the R square value and p-value. The low values concluded that the linear regression model is not a good fit model for this prediction. This report used linear regression analysis and hypothesis testing for determining the considered relationship (i.e., between open minded thinking and general conspiracist thinking). The findings have supported that open minded people are more likely to engage in general conspiracist thinking.

Therefore, it is recommended to take actions and prevent the spread of misinformation, and it can be done by increasing awareness amount the public and educating people to evaluate the information shared on social media.

Are you struggling to keep up with the demands of your academic journey? Don't worry, we've got your back!
Exam Question Bank is your trusted partner in achieving academic excellence for all kind of technical and non-technical subjects. Our comprehensive range of academic services is designed to cater to students at every level. Whether you're a high school student, a college undergraduate, or pursuing advanced studies, we have the expertise and resources to support you.

To connect with expert and ask your query click here Exam Question Bank