Data analysis and interpretation

Research hypotheses

The following alternative hypotheses were tested:

Alternative hypotheses

i) There is a statistically significant relationship between gender and how often do the respondents come across news stories about politics and government online that they think are not fully accurate.

ii) There is a statistically significant relationship between gender and how confident they are in their own ability to recognize news that is made up.

iii) There is a statistically significant relationship between the party and whether the respondents have ever shared a political news story online that they later found out was made up.

iv) There is a statistically significant relationship between the level of education and whether they have ever shared a political news story online that they thought at the time was made up.

Null hypotheses

The following null hypotheses were tested:

i) There is no statistically significant relationship between gender and how often do the respondents come across news stories about politics and government online that they think are not fully accurate.

ii) There is no statistically significant relationship between gender and how confident they are in their own ability to recognize news that is made up.

iii) There is no statistically significant relationship between the party and whether the respondents have ever shared a political news story online that they later found out was made up.

iv) There is no statistically significant relationship between the level of education and whether they have ever shared a political news story online that they thought at the time was made up.

Results

Descriptive statistics

The findings and interpretation of the study have been represented in this section. The results have been sub-divided into sections and subsections. The demographic information of the respondents such as gender, age, level of education, family level of income and the party has been presented first, (Bricker, et, al, 2017). After the demographic findings of the study have been discussed the researcher presented the research findings based on the study hypotheses. 

The quantitative data were analyzed using both descriptive and inferential statistics. The descriptive statistics were used to describe and summarize the data in form of charts, tables, frequencies, and percentages, (Anderson, Sweeney, Williams, Camm, & Cochran, 2016). The inferential statistics were used to help make inferences and draw conclusions, (Mertler, & Reinhart, 2016). Statistical tests including independent sample t-tests analysis, chi-square tests, and Spearman correlation were used to test the hypotheses. 

An independent sample t-test was used to establish whether there was a statistically significant relationship between gender and how often do the respondents come across news stories about politics and government online that they think are not fully accurate and whether there is a statistically significant relationship is no statistically significant relationship between gender and how confident they are in their own ability to recognize news that is made up.

In addition, the chi-square test was used to test to determine whether there was any statistically significant influence of party on whether the respondents have ever shared a political news story online that they later found out was made up. 

Finally, Spearman correlation was used to test whether there is a statistically significant relationship between the level of education and whether they have ever shared a political news story online that they thought at the time was made up. All tests of significance were computed at α = 0.05. The Statistical Package for Social Sciences (SPSS) version 22 was used to analyze the data.

Demographic information of the respondents

The data used in this study was drawn from the population of 728 respondents. The demographic characteristics of the respondents were summarized below.

Table 1 : Response by sex

	Frequency	Percent
Male	361	49.6
Female	367	50.4
Total	728	100.0

Table 1 shows the percentage distribution of the respondents in terms of their gender. The female participants were 367 (50.4%) which was slightly more than the male participants who were 361 (49.6%). From the analysis, the findings indicate that on average the female respondents were generally slightly more than their male counterparts.

Figure 1 : Age categories

According to the findings in Figure 1, the results showed that majority of the respondents were 60 years and above; 287 (39.4%). This was followed by 50-59 years old respondents at 153 (21.0%). In addition, 20-29 years and 30-39 years old were 93 (12.8%) and 89 (12.2%) respectively. Furthermore, 40-49 years respondents accounted for 85 (11.7%) while 21 (2.9%) were respondents aged from 10-19 years old. The mean age recorded was 52.73 years with a standard deviation of 18.80 while the range was between 18 years and 96 years.

The findings in Table 2 indicates that majority of the respondents; 187 (25.7%), 177 (24.3%0, 115 (15.8%) and 107 (14.7%) had High school graduate (Grade 12 with diploma or GED certificate), Four-year college or university degree/Bachelor's degree (e.g., BS, BA, AB), Postgraduate or professional degree, including master's, doctorate, medical or law degree (e.g., MA, MS, PhD, MD, JD) and Some college, no degree (includes some community college) respectively.

Table 2 : Level of education

Level of education	Frequency	Percent
Less than high school (Grades 1-8 or no formal schooling)	19	2.6
High school incomplete (Grades 9-11 or Grade 12 with NO diploma)	30	4.1
High school graduate (Grade 12 with diploma or GED certificate)	187	25.7
Some college, no degree (includes some community college)	107	14.7
Two year associate degree from a college or university	80	11
Four year college or university degree/Bachelor's degree (e.g., BS, BA, AB)	177	24.3
Some postgraduate or professional schooling, no postgraduate degree	13	1.8
Postgraduate or professional degree, including master's, doctorate, medical or law degree (e.g., MA, MS, PhD, MD, JD)	115	15.8
Total	728	100

The results in Table 3 shows that on average, most of the respondents had afamily income of less than $10,000 to $150,000 or over without a very big margin. For instance, 86 (11.8%) and 82 (11.3%) indicated their level of income to be between $75,000 to under $100,000 and $150,000 or over respectively.

Table 3 : Level of family income before taxation

Family income before taxation	Frequency	Percent
Less than $10,000	69	9.5
$10,000 to under $20,000	92	12.6
$20,000 to under $30,000	67	9.2
$30,000 to under $40,000	75	10.3
$40,000 to under $50,000	81	11.1
$50,000 to under $75,000	98	13.5
$75,000 to under $100,000	86	11.8
$100,000 to under $150,000	78	10.7
$150,000 or over	82	11.3
Total	728	100

Figure 2 : Respondent's party

On the party of the respondents, the results in Figure 2 indicates that 270 (37.1% and 261 (35.9%) of the respondents belongs to Democrat and Independent parties respectively while 197 (27.1%) were from the Republican party.

The findings in Table 4 shows that majority of the respondents; 276 (37.9%) believed that they have moderate views when it comes to politics while 229 (31.5%) were conservative. On the other hand, 127 (17.4%) and 52 (7.1%) were liberal or very liberal respectively on the political issues.

Table 4 : Respondent's political views

Respondent’s political views	Frequency	Percent
Very conservative	44	6.0
Conservative	229	31.5
Moderate	276	37.9
Liberal [OR]	127	17.4
Very liberal	52	7.1
Total	728	100.0

Hypotheses testing

The results in Table 5 shows that majority of the respondents; 398 (54.7%) and 196 (26.9%) do often or sometimes respectively come across news stories about politics and government online that they think are not fully accurate.

Table 5 : News that may not be fully accurate

	Frequency	Percent
Never	73	10.0
Hardly ever	61	8.4
Sometimes	196	26.9
Often	398	54.7
Total	728	100.0

An independent sample t-tests analysis was used to test whether there was a statistically significant relationship between gender and how often do the respondents come across news stories about politics and government online that they think are not fully accurate. According to the results, on average the female score (M=2.24, SE=0.053), was not significantly lower than the male score (M= 2.29, SE=0.050), t (726) = 0.701, p = .407. Therefore, the study concludes that there is no statistically significant relationship between gender and how often do the respondents come across news stories about politics and government online that they think are not fully accurate, ( Allcott, & Gentzkow, 2017) .

The results in Table 6 shows that majority of the respondents; 305 (41.9%) and 234 (32.1%) somewhat confident and very confident respectively that they are confident they are in their own ability to recognize news that is made up.

Table 6 : Completely made political stories

Political stories completely made up	Frequency	Percent
Not at all confident	78	10.7
Not very confident	111	15.2
Somewhat confident	305	41.9
Very confident	234	32.1
Total	728	100.0

An independent sample t-tests analysis was used to test whether there is a statistically significant relationship between gender and how confident they are in their own ability to recognize news that is made up. According to the results, on average the female score (M=1.91, SE=0.05), was not significantly lower than the male score (M= 2.00, SE=0.49), t (726) = 1.271, p = .618. Therefore, the study concludes that there is no statistically significant relationship between gender and how confident they are in their own ability to recognize news that is made up.

Table 7 : Political news shared and were made up

	Frequency	Percent
Yes	126	17.3
No	602	82.7
Total	728	100.0

On whether the respondents have ever shared a political news story online that they later found out was made up, 602 (82.7%) sated no meaning that they have never while only 126 (17.3%) in one way or the other have ever shared the political news that later found to be made up. According to the chi-square results, the Pearson chi-square showed that there is no statistically significant relationship between the party and whether the respondents have ever shared a political news story online that they later found out was made up, (chi =2.577, p=0.276).

Table 8 : Shared political stories at the time it was made up

	Frequency	Percent
Yes	112	15.4
No	616	84.6
Total	728	100.0

Finally, the findings in Table 8 shows that 112 (15.4%) have ever shared a political news story online that they thought at the time was made up while 616 (84.6%) have not. The Pearson Product-Moment correlation coefficient (r = .399, p= 0.031) computed indicated that there was a high positive correlation between the level of education and whether the respondents have ever shared a political news story online that they thought at the time was made up, (Nardi, 2018).

References

Allcott, H., & Gentzkow, M. (2017). Social media and fake news in the 2016 election.  Journal of Economic Perspectives ,  31 (2), 211-36.

Anderson, D. R., Sweeney, D. J., Williams, T. A., Camm, J. D., & Cochran, J. J. (2016).  Statistics for business & economics . Nelson Education.

Bricker, J., Dettling, L. J., Henriques, A., Hsu, J. W., Jacobs, L., Moore, K. B., ... & Windle, R. A. (2017). Changes in US family finances from 2013 to 2016: evidence from the survey of consumer finances.  Fed. Res. Bull. ,  103 , 1.

Mertler, C. A., & Reinhart, R. V. (2016).  Advanced and multivariate statistical methods: Practical application and interpretation . Routledge.

Nardi, P. M. (2018).  Doing survey research: A guide to quantitative methods . Routledge.