IPSOS / REUTERS POLL DATA Prepared by Ipsos Public Affairs Ipsos Poll Conducted for Reuters Apple and the FBI Topline 2.23.2016 These are findings from an Ipsos poll conducted February 19-23, 2016 on behalf Thomson Reuters. For the survey, a sample of roughly 1,576 adults age 18+ from the continental U.S., Alaska and Hawaii was interviewed online in English. The sample for this study was randomly drawn from Ipsos’s online panel (see link below for more info on “Access Panels and Recruitment”), partner online panel sources, and “river” sampling (see link below for more info on the Ipsos “Ampario Overview” sample method) and does not rely on a population frame in the traditional sense. Ipsos uses fixed sample targets, unique to each study, in drawing sample. After a sample has been obtained from the Ipsos panel, Ipsos calibrates respondent characteristics to be representative of the U.S. Population using standard procedures such as raking-ratio adjustments. The source of these population targets is U.S. Census 2015 American Community Survey data. The sample drawn for this study reflects fixed sample targets on demographics. Post-hoc weights were made to the population characteristics on gender, age, region, race/ethnicity and income. Statistical margins of error are not applicable to online polls. All sample surveys and polls may be subject to other sources of error, including, but not limited to coverage error and measurement error. Where figures do not sum to 100, this is due to the effects of rounding. The precision of Ipsos online polls is measured using a credibility interval. In this case, the poll has a credibility interval of plus or minus 2.8 percentage point for all respondents (see link below for more info on Ipsos online polling “Credibility Intervals”). Ipsos calculates a design effect (DEFF) for each study based on the variation of the weights, following the formula of Kish (1965). This study had a credibility interval adjusted for design effect of the following (n=1,576, DEFF=1.5, adjusted Confidence Interval=4.3). For more information about Ipsos online polling methodology, please go here http://goo.gl/yJBkuf TM53Y13_1_1 - Would you be willing to Give up privacy of my email if it would...Help the US government foil foreign terrorist plots? *Added on 2/18/2016 No Yes Total Total 69% 31% 1576 TM53Y13_1_2 - Would you be willing to Give up privacy of my email if it would...Help the US government foil domestic terrorist plots? *Added on 2/18/2016 No Yes Total 71% 29% 1576 TM53Y13_1_3 - Would you be willing to Give up privacy of my email if it would...Help the US government counter hacking of US networks and infrastructure by foreign powers? *Added on 2/18/2016 No Yes 78% 22% Total 1576 TM53Y13_1_4 - Would you be willing to Give up privacy of my email if it would...None of these? *Added on 2/18/2016 No Yes Total 41% 59% 1576 TM53Y13_2_1 - Would you be willing to Give up privacy of my text messages if it would...Help the US government foil foreign terrorist plots? *Added on 2/18/2016 No Yes 75% 25% Total 1576 No Yes 72% 28% TM53Y13_2_2 - Would you be willing to Give up privacy of my text messages if it would...Help the US government foil domestic terrorist IPSOS / REUTERS POLL DATA Prepared by Ipsos Public Affairs plots? *Added on 2/18/2016 Total 1576 No Yes 79% 21% Total 1576 TM53Y13_2_4 - Would you be willing to Give up privacy of my text messages if it would...None of these? *Added on 2/18/2016 No Yes Total 38% 62% 1576 TM53Y13_3_1 - Would you be willing to Give up privacy of my phone records if it would...Help the US government foil foreign terrorist plots? *Added on 2/18/2016 No Yes 71% 29% Total 1576 No Yes 70% 30% Total 1576 No Yes 76% 24% Total 1576 TM53Y13_3_4 - Would you be willing to Give up privacy of my phone records if it would...None of these? *Added on 2/18/2016 No Yes Total 42% 58% 1576 TM53Y13_4_1 - Would you be willing to Give up privacy of my internet activities if it would...Help the US government foil foreign terrorist plots? *Added on 2/18/2016 No Yes 74% 26% Total 1576 No Yes 73% 27% Total 1576 No Yes 77% 23% Total 1576 TM53Y13_4_4 - Would you be willing to Give up privacy of my internet activities if it would...None of these? *Added on 2/18/2016 No Yes Total 38% 62% 1576 TM853Y16 - Apple is opposing a court order to unlock a smart phone that was used by one of the shooters in the San Bernardino attack. Apple is concerned that if it helps the FBI this time, it will be forced to help the government in future cases that may not be linked to national security, opening the door for hackers and potential future Agree 46% Disagree 35% Don't know 20% TM53Y13_2_3 - Would you be willing to Give up privacy of my text messages if it would...Help the US government counter hacking of US networks and infrastructure by foreign powers? *Added on 2/18/2016 TM53Y13_3_2 - Would you be willing to Give up privacy of my phone records if it would...Help the US government foil domestic terrorist plots? *Added on 2/18/2016 TM53Y13_3_3 - Would you be willing to Give up privacy of my phone records if it would...Help the US government counter hacking of US networks and infrastructure by foreign powers? *Added on 2/18/2016 TM53Y13_4_2 - Would you be willing to Give up privacy of my internet activities if it would...Help the US government foil domestic terrorist plots? *Added on 2/18/2016 TM53Y13_4_3 - Would you be willing to Give up privacy of my internet activities if it would...Help the US government counter hacking of US networks and infrastructure by foreign powers? *Added on 2/18/2016 IPSOS / REUTERS POLL DATA Prepared by Ipsos Public Affairs data breaches for smartphone users. Do you agree or disagree with Apple’s decision to oppose the court order? Total 1284 TM854Y16_1 - Level of agreement with the following...The government should be able to look at data on Americans' phones in order to protect against terror threats Strongly agree Somewhat agree Somewhat disagree Strongly disagree Not sure Total 18% 28% 17% 25% 13% 1284 TM854Y16_2 - Level of agreement with the following...If Apple unlocks the iPhone for the government, cybercriminals will soon be able to steal data from locked iPhones Strongly agree Somewhat agree Somewhat disagree Strongly disagree Not sure Total 27% 27% 14% 12% 20% 1284 TM854Y16_3 - Level of agreement with the following...The government will use this ability to unlock phones to spy on iPhone users Strongly agree Somewhat agree Somewhat disagree Strongly disagree Not sure Total 30% 25% 13% 15% 17% 1284 IPSOS / REUTERS POLL DATA Prepared by Ipsos Public Affairs How to Calculate Bayesian Credibility Intervals The calculation of credibility intervals assumes that Y has a binomial distribution conditioned on the parameter θ\, i.e., Y θ~Bin(n,θ), where n is the size of our sample. In this setting, Y counts the number of “yes”, or “1”, observed in the sample, so that the sample mean (y ̅) is a natural estimate of the true population proportion θ. This model is often called the likelihood function, and it is a standard concept in both the Bayesian and the Classical framework. The Bayesian 1 statistics combines both the prior distribution and the likelihood function to create a posterior distribution. The posterior distribution represents our opinion about which are the plausible values for θ adjusted after observing the sample data. In reality, the posterior distribution is one’s knowledge base updated using the latest survey information. For the prior and likelihood functions specified here, the posterior distribution is also a beta distribution (π(θ/y)~β(y+a,n-y+b)), but with updated hyper-parameters. Our credibility interval for θ is based on this posterior distribution. As mentioned above, these intervals represent our belief about which are the most plausible values for θ given our updated knowledge base. There are different ways to calculate these intervals based on π(θ/y). Since we want only one measure of precision for all variables in the survey, analogous to what is done within the Classical framework, we will compute the largest possible credibility interval for any observed sample. The worst case occurs when we assume that a=1 and b=1 and y=n/2. Using a simple approximation of the posterior by the normal distribution, the 95% credibility interval is given by, approximately: For this poll, the Bayesian Credibility Interval was adjusted using standard weighting design effect 1+L=1.3 to account for complex weighting2 Examples of credibility intervals for different base sizes are below. Ipsos does not publish data for base sizes (sample sizes) below 100. Sample size 2,000 1,500 1,000 750 500 350 200 100 Credibility intervals 2.5 2.9 3.5 4.1 5.0 6.0 7.9 11.2