FINAL EXAM – PM SOCIAL RESEARCH METHODS
Faculty of Social Sciences, Vrije Universiteit Amsterdam
COURSE: PM_SRM_2017_1 (PM Social Research Methods)
DATE: Friday 16 February 2018
TIME: 18:30 – 21:15
LOCATION: MF-FG1 / MF-FG2
1. The exam consists of 34 multiple choice questions.
2. Read the multiple choice questions slowly and carefully, and always choose the BEST alternative.
There is only one correct answer.
3. Please place your student card on the desk in front of you.
4. Asking the invigilators to clarify test questions serves no purpose.
5. You have up to 2 hours and 45 minutes to answer the questions.
6. The use of external resources is PROHIBITED. Mobile telephones must be switched off (not on
silent mode, but off!) and stowed in a bag. Bags must be closed and may not be opened during
the examination. If it is found that you have broken this rule, you will immediately be
disqualified from the examination. The only external material allowed is an English-Dutch
dictionary.
7. When the result of the examination is known, you will have the opportunity to inspect the
graded examination. The date, time and location for this inspection will be announced on
Canvas.
Student number
First name
Between letters
Last name
GOOD LUCK!
1
,Question 1. Which of the following statements about the Likert scale is/are NOT correct? A
Likert scale...
a). Consists of multiple, different items.
b). Contains multiple, different indicators.
c). Measures multiple, different hypotheses.
d). Has multiple answer categories to capture the intensity of an attitude.
Question 2. What is the cause of making the so-called ‘ecological fallacy’?
This error occurs if we ...
a). Have a low sampling fraction.
b). Derive a concrete hypothesis from a general theory.
c). Refute an empirical observation because it is in conflict with the theory.
d). Derive a conclusion about the micro level from macro-level data.
Question 3. In one of the lectures the following picture was shown to explain the difference
between reliability and validity. Which of the following four measurements is/are reliable?
a). Only measurement 4 is reliable.
b). Measurements 2 and 4 are reliable.
c). Measurements 3 and 4 are reliable.
d). Measurements 2, 3 and 4 are reliable.
2
, Read the (abridged) article below. Questions 4 to 6 relate to this article.
Facebook influenced voter turnout
In the run-up to the US presidential election of 2012, Facebook presented news stories in such
a way that more of its users would vote. This is not the first secret experiment that Facebook
has carried out involving its users. Political experiments have also been done on Twitter. The
controversy has been significant.
Normally, news stories that have been shared by friends are buried in the continuous flow of all
the other items shown on the screen of Facebook users. However, Facebook analysts decided
to draw more attention to current events by placing news stories at the top. They wanted to see
if this would increase people’s propensity to vote.
And indeed, a survey of 1.9 million US Facebook users shows that there was a clear increase in
turnout. In a group of Facebook users, voter turnout rose from 64 percent to 67 percent.
According to a Facebook data researcher, the news stories led to a more socially engaged
attitude and this led to a higher turnout on election day. The latter was measured using a button
that allowed users to indicate whether they had voted.
Susceptible to manipulation
News of the experiment was revealed on Friday by Mother Jones, an American magazine.
Critics see it as fresh evidence that social media sites are very sensitive to manipulation. ‘If I
hadn’t stopped using Facebook, I would stop now,’ was one response on Twitter.
In a presentation of the results, which has now been removed from YouTube, Facebook data
analysts referred to their trick as a ‘treatment’ of Facebook users.
Facebook has been trying to boost turnout at elections since 2008. It introduced a digital ‘I
voted’ sticker in 2010, which was meant to encourage users’ Facebook friends to vote too.
According to research published in the scientific journal Nature, the social pressure led to
340,000 additional Americans voting, which means that Facebook could have been responsible
for 0.14 of the 0.6 percentage point increase in voter turnout.
Liberal, left-leaning
Because Facebook users tend to be more liberal and left-wing than the average American, the
increased turnout would tend to favour the Democrats. Facebook commented that ‘voting is a
core value of democracy that we would like to encourage’.
Source: De Volkskrant, 4 November 2014
3
Faculty of Social Sciences, Vrije Universiteit Amsterdam
COURSE: PM_SRM_2017_1 (PM Social Research Methods)
DATE: Friday 16 February 2018
TIME: 18:30 – 21:15
LOCATION: MF-FG1 / MF-FG2
1. The exam consists of 34 multiple choice questions.
2. Read the multiple choice questions slowly and carefully, and always choose the BEST alternative.
There is only one correct answer.
3. Please place your student card on the desk in front of you.
4. Asking the invigilators to clarify test questions serves no purpose.
5. You have up to 2 hours and 45 minutes to answer the questions.
6. The use of external resources is PROHIBITED. Mobile telephones must be switched off (not on
silent mode, but off!) and stowed in a bag. Bags must be closed and may not be opened during
the examination. If it is found that you have broken this rule, you will immediately be
disqualified from the examination. The only external material allowed is an English-Dutch
dictionary.
7. When the result of the examination is known, you will have the opportunity to inspect the
graded examination. The date, time and location for this inspection will be announced on
Canvas.
Student number
First name
Between letters
Last name
GOOD LUCK!
1
,Question 1. Which of the following statements about the Likert scale is/are NOT correct? A
Likert scale...
a). Consists of multiple, different items.
b). Contains multiple, different indicators.
c). Measures multiple, different hypotheses.
d). Has multiple answer categories to capture the intensity of an attitude.
Question 2. What is the cause of making the so-called ‘ecological fallacy’?
This error occurs if we ...
a). Have a low sampling fraction.
b). Derive a concrete hypothesis from a general theory.
c). Refute an empirical observation because it is in conflict with the theory.
d). Derive a conclusion about the micro level from macro-level data.
Question 3. In one of the lectures the following picture was shown to explain the difference
between reliability and validity. Which of the following four measurements is/are reliable?
a). Only measurement 4 is reliable.
b). Measurements 2 and 4 are reliable.
c). Measurements 3 and 4 are reliable.
d). Measurements 2, 3 and 4 are reliable.
2
, Read the (abridged) article below. Questions 4 to 6 relate to this article.
Facebook influenced voter turnout
In the run-up to the US presidential election of 2012, Facebook presented news stories in such
a way that more of its users would vote. This is not the first secret experiment that Facebook
has carried out involving its users. Political experiments have also been done on Twitter. The
controversy has been significant.
Normally, news stories that have been shared by friends are buried in the continuous flow of all
the other items shown on the screen of Facebook users. However, Facebook analysts decided
to draw more attention to current events by placing news stories at the top. They wanted to see
if this would increase people’s propensity to vote.
And indeed, a survey of 1.9 million US Facebook users shows that there was a clear increase in
turnout. In a group of Facebook users, voter turnout rose from 64 percent to 67 percent.
According to a Facebook data researcher, the news stories led to a more socially engaged
attitude and this led to a higher turnout on election day. The latter was measured using a button
that allowed users to indicate whether they had voted.
Susceptible to manipulation
News of the experiment was revealed on Friday by Mother Jones, an American magazine.
Critics see it as fresh evidence that social media sites are very sensitive to manipulation. ‘If I
hadn’t stopped using Facebook, I would stop now,’ was one response on Twitter.
In a presentation of the results, which has now been removed from YouTube, Facebook data
analysts referred to their trick as a ‘treatment’ of Facebook users.
Facebook has been trying to boost turnout at elections since 2008. It introduced a digital ‘I
voted’ sticker in 2010, which was meant to encourage users’ Facebook friends to vote too.
According to research published in the scientific journal Nature, the social pressure led to
340,000 additional Americans voting, which means that Facebook could have been responsible
for 0.14 of the 0.6 percentage point increase in voter turnout.
Liberal, left-leaning
Because Facebook users tend to be more liberal and left-wing than the average American, the
increased turnout would tend to favour the Democrats. Facebook commented that ‘voting is a
core value of democracy that we would like to encourage’.
Source: De Volkskrant, 4 November 2014
3
