Fact-checking resources: A guide to finding reliable answers to timely questions

How have immigration rates changed for unaccompanied minors at the U.S. southwest border, and what countries are the children coming from?

You can find 2014 vs. 2013 comparison rates here and here.

TIP: The breakdown by country shows the origin of about 98 percent of unaccompanied children apprehended at the southwest border (by API’s calculation).

Note that while the number of unaccompanied children has near doubled year-on-year (as shown in the two links above), the number has also been increasing since FY 2011.

What happens to unaccompanied children that cross over the border into the U.S.?

See this Health and Human Services fact sheet.

How has the number of undocumented immigrants residing in the U.S. changed, which regions/countries have most of these immigrants come from and what state do most end up in?

Department of Homeland Security estimates.

For state of residence, a Pew report has slightly different estimates than the DHS report above, and it shows longer-term trends for the six states with the highest number of unauthorized immigrants: California, Florida, Illinois, New Jersey, New York and Texas.

How have rates of apprehension for illegal immigrants changed, and what regions/countries do most of these people come from?

See U.S. Border Patrol apprehensions by year; and apprehensions by month and border/sector.

Also see the Department of Homeland Security’s 2012 Yearbook of Immigration Statistics.

TIP: The Border Patrol’s yearly figures go back to 1925, but the monthly figures go back only to FY 1999. 

Total illegal apprehension statistics from the DHS include not just apprehensions by the Border Patrol, but also by two divisions of U.S. Immigrations and Customs Enforcement: Homeland Security Investigations, and Enforcement and Removal. In the DHS yearbook, Table 33 shows alien apprehensions by year, FY 1925 to FY 2012. Table 34 has a breakdown by all regions and countries, from 2003 to 2012.

In addition, Border Patrol spokesman Doug Mosier provided PolitiFact Texas with this chart showing monthly apprehensions along the U.S.-Mexico border through early September 2014.

How has the number of foreign-born in the U.S. changed over the past five years?

See the U.S. Census’s American FactFinder.

TIP: Note that these directions yield data for the foreign-born population. The data for 2013, for example, shows all the foreign-born people living in the U.S. in that year – not the number of foreign-born people who immigrated to the U.S. in that year.

• Click Advanced Search, then Show Me All.
• Under “Select a geographic type,” select United States.
• In the next menu, again select United States.
• Then click “Add to your selection.”
• Then click the close button at the top-right of the Geographic Areas window.
• Next, in the tan “Refine your search results” area near the top, under “topic or table name,” type or paste this code: CP02. Hit enter.
• Click 2013 ACS 1-year estimates. This will give you the most up-to-date data.

This is a chart of many social characteristics – you have to scroll a bit to get to Place of Birth (which includes foreign-born). Scroll down a bit more and you will also see World Region of Birth.

Where do immigrants to the U.S. come from?

See Table 2.17 in this Census Bureau document or the U.S. Census’s American FactFinder.

TIP: Table 2.17 breaks down the foreign-born population according to their decade of entry and area of birth: Asia, Europe, Mexico, other Latin America, and other.

FactFinder shows data for the foreign-born population resident in the U.S., not for immigration. The data for 2013, for example, shows all the foreign-born people living in the U.S. – not the number of foreign-born people who immigrated to the U.S. in that year. FactFinder does, however, have a country-by-country breakdown.

• From the FactFinder homepage, click Advanced Search, then Show Me All.
• Under “Select a geographic type,” select United States.
• In the next menu, again select United States.
• Then click “Add to your selection.”
• Then click the close button at the top-right of the Geographic Areas window.
• Next, in the tan “Refine your search results” area near the top, under “topic or table name,” type or paste this code: B05006.
• Hit enter.
• Click 2013 ACS 1-year estimates. This will give you the most up-to-date data.

The Census Bureau says that this detailed level of data is not available as a time series. If you want to compare this data across years, you will have to download each year separately and manually combine the data. Alternatively, you can use the CPO2 data set (see How has the number of foreign-born in the U.S. changed over the past five years?, above), if you only need continent of origin and not country.

For more information, see the Pew Research Center fact sheet, “Unauthorized Immigrants: Who they are and what the public thinks.”

How has the number of foreign-born in my state or metro area changed over the past five years?

See the U.S. Census’s American FactFinder.

TIP: Note that these directions yield data for the foreign-born population. The data for 2013, for example, shows all the foreign-born people living in the state or metro area in that year – not the number of foreign-born people who immigrated to that area in that year.

• Click Advanced Search, then Show Me All.
• Then under “Select a geographic type,” select State, Metropolitan Statistical Area, or other area of interest.
• Then choose your state, metro area, etc. (You can also choose “All states” to see data for all states.)
• Then click “Add to your selection.”
• Then click the close button at the top-right of the Geographic Areas window.
• Next, in the tan “Refine your search results” area near the top, under “topic or table name,” type or paste this code: CP02. Hit enter.
• Click 2013 ACS 1-year estimates. This will give you the most up-to-date data.

This is a chart of many social characteristics – you have to scroll a bit to get to Place of Birth (which includes foreign-born). Scroll down a bit more and you will also see World Region of Birth.

Where did the foreign-born population in my state or metro area come from?

See the U.S. Census’s American FactFinder.

TIP: Note that these directions yield data for the foreign-born population. The data for 2013, for example, shows all the foreign-born people living in the state or metro area in that year – not the number of foreign-born people who immigrated to that area in that year.

• Click Advanced Search, then Show Me All.
• Then under “Select a geographic type,” select State, Metropolitan Statistical Area, or other area of interest.
• Then choose your state, metro area, etc. (You can also choose “All states” to see data for all states.)
• Then click “Add to your selection.”
• Then click the close button at the top-right of the Geographic Areas window.
• Next, in the tan “Refine your search results” area near the top, under “topic or table name,” type or paste this code: B05006. Hit enter.
• Click 2013 ACS 1-year estimates. This will give you the most up-to-date data.

The Census Bureau says that this detailed level of data is not available as a time series. If you want to compare this data across years, you will have to download each year separately and manually combine the data. Alternatively, you can use the CPO2 data set (see How has the number of foreign-born in my state or metro area changed over the past five years?, above), if you only need continent of origin and not country.

How many people have died trying to cross the US southern border?

US Border Patrol data shows Southwest Border deaths, 1998-2013.

What are the basics of the Affordable Care Act (also known as Obamacare)?

See Health Reform GPS.

For more on exemptions to the fee for the uninsured, see Healthcare.gov or the Kaiser Family Foundation.

The law’s text is available here.

What are the legislative challenges to ACA? 

See this September 2014 report from the Congressional Research Service.

Which states are refusing to expand Medicaid, and which ones are undecided?

Tracked by the Henry J. Kaiser Family Foundation.

How many have people signed up for health insurance through the Affordable Care Act exchanges?

See this U.S. Department of Health and Human Services report.

TIP: Reports are posted here periodically.

How many people have joined Medicaid since it expanded under Obamacare?

See this U.S. Department of Health and Human Services report.

TIP: Reports are posted here periodically.

How has the number of uninsured changed since Obamacare was introduced? 

See the Centers for Disease Control and Prevention’s National Health Interview Survey.

TIP: This edition of the report Health Insurance Coverage: Estimates from the National Health Interview Survey, published in September, is the first federal measure of the number of uninsured Americans since the Affordable Care Act took effect. The report is published each quarter.

There is some disagreement over the accuracy of these figures. Larry Levitt of the Kaiser Family Foundation said the CDC figures “dramatically understate” the effect of the law, with some surveys showing higher enrollment figures. Most importantly, the true effect of ACA will probably not be clear for several more months.

How have health insurance costs changed?

For data on employer-sponsored health insurance, see the Kaiser Family Foundation.

TIP: When trying to show changes in health insurance costs, proceed with caution. The best measure is probably employer-based policies for single-person and family premiums.

It’s hard to make comparisons for the individual market. Obamacare changed how insurers can price plans on the individual market, which was already volatile. Plans now have to include certain essential benefits that were not mandated before. The Kaiser Family Foundation has actually concluded that these comparisons are impossible.

And in fact, insurance premiums go up every year, so any increase – even if it were made on an apples-to-apples basis – would be difficult to pin on Obamacare.

Can Ebola be spread through a sneeze or cough? Can it go airborne?

For the Ebola virus’s current properties, see this statement and this one from the Centers for Disease Control and Prevention; commentary published by the University of Minnesota’s Center for Infectious Disease Research and Policy; and a subsequent CIDRAP statement.

TIP: Some (notably George Will ) have misinterpreted commentary from two University of Illinois professors, posted by the University of Minnesota’s Center for Infectious Disease Research and Policy. The piece argued that health care workers should wear respirators when treating patients with Ebola, to prevent inhaling any virus contained in vomit and diarrhea, when patients are in the most severe stage of the disease. This has nothing to do with the risk of infection from coughs and sneezes in public places.

Politifact cites Stephen Gire, a research scientist in the Sabeti Lab at Harvard University, who said that to get Ebola from a cough or sneeze, the infected person would need to cough or sneeze on you directly, and droplets from this cough or sneeze would have to be transferred into your mucosal membranes. This scenario is very unlikely, Gire says. He also notes that this form of transmission counts as direct contact, and does not mean the virus is “airborne.”

According to this FAQ from Mount Sinai Hospital’s Department of Microbiology, “airborne transmission” refers to situations where dust particles or droplet residue containing microorganisms remain suspended in the air for a long period. The organisms must be resistant to drying, and must be able to survive outside the body for a long time.

As to whether the virus can mutate and go airborne, the CDC says there is no evidence Ebola is mutating in this way. In interviews with Vox  and Scientific American, virologists said the chance of this happening is remote, for many reasons. For one thing, we have never seen a human virus change its mode of transmission. But Michael T. Osterholm, director of the Center for Infectious Disease Research and Policy at the University of Minnesota, provides a note of dissension. In a New York Times op-ed, he said the possibility of Ebola going airborne “is one that virologists are loath to discuss openly but are definitely considering in private.”

With thanks to PolitiFact. 

Is this disease on the rise? How prevalent is this disease in my community? 

See the Centers for Disease Control and Prevention’s Morbidity and Mortality Weekly Report; Health Indicators Warehouse ; CDC Wonder; CDC A-Z Index; Health Data Interactive; National Health Interview Survey; Health, United States;   Surveillance Resource Center; and National Center for Health Statistics.

TIP: For infectious diseases, the best places to start are the Morbidity and Mortality Weekly Report and Health Indicators Warehouse.

In the Morbidity and Mortality Weekly Report, look especially at the National Notifiable Diseases Surveillance System (NNDSS) Tables for morbidity. These present data for selected nationally notifiable diseases reported by the 50 states, New York City, the District of Columbia, and U.S. territories, collated and published weekly. Table 1 provides yearly (including year-to-date) data for over 40 diseases, a slightly larger number than the weekly tables.

For deaths from influenza and pneumonia, look at the NNDSS Mortality Tables, which present number of deaths each week in 122 cities and metropolitan areas, reported within three weeks from the date of death.

The Health Indicators Warehouse offers data on chronic diseases, including arthritis, back conditions, cancer, diabetes, heart disease, stroke, kidney disease, osteoperosis, respiratory disease and venous thromboembolism; and infectious diseases, such as food and waterborne diseases, AIDS and other STDs, flu, hepatitis, malaria, meningits and tuberculosis; as well as many other health indicators.

Another good resource is CDC Wonder. You can search for a specific term, use its A-Z index or browse by topic. Some of the data available on Wonder includes leading causes of death, cancer incidence and mortality, diabetes data and an STD atlas.

The CDC’s A-Z index is also a good starting point for many diseases. Go to the CDC homepage and click “CDC A-Z.” Then use the index to find the disease you’re interested in. From there, look for “Maps and Data” and “Statistical Reports.”

Health Data Interactive covers ADHD, allergies, arthritis, asthma, cancer, diabetes, heart disease, high cholesterol, hypertension, incontinence, obesity, psychological distress, stroke, tooth decay and tooth loss. Tables can be customized by geographic location as well as well as age, gender and race/ethnicity. A tutorial is available here. 

The National Health Interview Survey covers obesity, diabetes and asthma, along with more generalized health issues like vaccination, insurance status and health-related behaviors.

Finally, the CDC’s Health, United States report gives annual data on the prevalence of a few major diseases, such as asthma, ADHD, hypertension, diabetes and heart disease.

Further health statistics resources are available from the CDC’s Surveillance Resource Center and National Center for Health Statistics.

Is it really that important to get vaccinated against the flu? 

See CDC fact sheets 2014-2015 Seasonal Influenza Vaccine Safety, Vaccination: Who Should Do It, Who Should Not and Who Should Take Precautions, and Vaccine Effectiveness: How Well Does the Flu Vaccine Work?

TIP: The short answer is: yes. Most populations are likely to benefit, and few are at risk of harm. The CDC found that flu vaccinations in the 2013-2014 season prevented an estimated 7.2 million flu-associated illnesses and 90,000 flu hospitalizations, cutting both illness and hospitalization 17 percent from what would have occurred without vaccinations.

The benefits of the vaccine outweigh the risk, for most people. Getting the flu is just a discomfort for many, but these healthy adults can pass the disease to at-risk people for whom the flu can be deadly. The CDC estimates annual flu deaths ranged from 3,000 to 49,000 people between 1976 and 2007. Although it’s possible to get the flu even after a shot, vaccination can make the illness milder.

For most people, the risks from flu vaccination are minor. These include mild fever and aches, and in the case of the nasal vaccine, runny nose, headache, sore throat and coughing. On rare occasions, people can have severe allergic reactions to flu vaccination.

For these reasons, the CDC recommends that everyone 6 months of age and older get the flu shot every year, with very few exceptions.

How much of a problem is traumatic brain injury among children and teenagers? 

See Nonfatal Traumatic Brain Injuries Related to Sports and Recreation Activities Among Persons Aged ≤19 Years.

TIP: This 2011 study published in the CDC’s Morbidity and Mortality Weekly Report found that in 2009, about 248,418 children, age 19 or younger, were treated in U.S. emergency departments for sports and recreation-related injuries with a diagnosis of concussion or TBI. And from 2001 to 2009, the rate of such visits among children rose 57 percent.

With thanks to Journalist’s Resource.

How has the unemployment rate changed?

The U.S. Bureau of Labor Statistics offers several measures of unemployment, based on the CPS. (See above for explanation of CPS vs. CES.)

TIP: The Bureau of Labor Statistics provides several different measures of unemployment. The “official rate,” which is the most commonly cited one, divides the country into:

• Employed
• Unemployed – those who are available to work and have actively sought a job in the past four weeks
• Not in the labor force – this includes retirees, children and full-time students, but also jobless people who have not actively sought a job in the past four weeks.

Other BLS unemployment and underemployment rates account for measures such as jobless people too discouraged to seek work, and those who want full-time work but have had to settle for part-time.

When fact-checking, it is important to note which unemployment rate is being used.

There’s a good overview of some of the different CPS measures here.

How has employment changed nationally?

See the Current Employment Statistics data.

TIP: Click Multi-Screen, then seasonally adjusted, total nonfarm, all employees, and total nonfarm again.

What have the job growth patterns been nationally? 

See the Current Employment Statistics data.

How has employment changed in my state or metro area?

See CES data here. (See above for explanation of CPS vs. CES.)

TIP: Probably the easiest way to use this is the multi-screen option: Look for the yellow button, or access directly here. More details on how to use the databases here. Note that this data draws on the Current Employment Statistics (CES) program, so if you wish to compare to national trends, you’ll need to use national CES (rather than CPS) data. This can be found here (or go directly to the multi-screen option here). 

How has the consumer price index changed, by month and year?

See the Consumer Price Index for All Urban Consumers.

TIP: This index covers about 87 percent of the U.S. population (source).

How much would this good or service cost in “today’s money?”

See the CPI Inflation Calculator.

What is the average income in my state, and how has this changed? 

See U.S. Bureau of Economic Analysis figures.

TIP: Select Real Personal Income and Regional Price Parities, then Real Personal Income.

• Choose your desired area.
• Leave on “Levels” if you want to see income levels, or “Percent change from preceding period” if you want to see a growth chart
• Select “Real per capita personal income.”
• Select desired years. The data goes back to 2008.

How much will this regulation cost the economy?

You can search for economically significant regulations here, and use the Regulatory Review Dashboard to view all regulations of economic significance currently being reviewed by the Office of Information and Regulatory Affairs.

TIP: OIRA decides whether proposed regulations are “significant.” One way a regulation gets designated as “significant” is if its projected economic effect is $100 million or more. OIRA designates about 500-700 regulatory actions as significant each year, and is charged with coordinating executive branch reviews of significant regulations.

Who has contributed to this Congressional campaign, and how much did they give?

Check out MapLight.

TIP: The website notes: “MEDIA: Due to the nature of campaign finance data there may be multiple spellings of an organization or individual. To ensure the accuracy of your findings please contact the MapLight research department via e-mail or call 510-868-0894 to verify totals before publishing a citation.”

How much has each candidate in this Congressional race raised? How much have outside groups (PACs, SuperPACs, 501(c)s, etc.) spent in this campaign?

Check out OpenSecrets.org.

TIP: Click your state or choose it from the drop-down menu.

First you’ll see a summary listing each race with its candidates and the total each has raised.

Click the race name for more detailed results. Even more detailed results can be viewed by clicking on the tabs “Candidates,” “Contributors” and so forth.

Which industries and individual contributors gave this Congressional campaign the most money?

Check out Influence Explorer.

TIP: Search for candidate, then click on his/her name on results page.

How did this member of Congress vote on a particular bill?

Check out Open Congress.

TIP: Enter the congress member’s name in the search field. On the results page, click on the congress member’s name for a more detailed page on his/her record. In the tabs (“Overview” is selected by default), click “Votes.” This lists votes in reverse chronological order. Enter a particular bill number in search field, and you will see all the person’s votes on that bill. You can also enter a search term, though this will of course be a more scattershot approach.

What bills has this member of Congress sponsored? 

Check out Open Congress.

TIP: Enter the member’s name in the search field.

On the results page, click on the member’s name for a more detailed page on his/her record.

This will show the number of bills he/she sponsored or co-sponsored this session.

Scroll down to click on “View All Sponsored Bills” and “View All Co-Sponsored Bills” buttons.

How often has this member of Congress voted with his/her party?

Check out Open Congress.

TIP: Enter the member’s name in the search field.

On the results page, click on the member’s name for a more detailed page on his/her record, including percentage of votes with the member’s party.

What committees does this member of Congress sit on?

Check out Open Congress.

TIP: Enter the member’s name in the search field.

On the results page, click on the member’s name for a more detailed page, including his/her committee memberships.

How did this state senator or state representative vote on a particular bill? How much did he/she raise in each campaign cycle?

Check out Open States.

TIP: This website shows each legislator’s year-by-year campaign revenues, sponsored bills, committee membership and recent votes.

Who is this political group? Where did they get their funding? And who are they supporting? 

See OpenSecrets.org.

TIP: Type the group’s name in the search bar and hit the search button.

On the results page, the first tab will show matches for that organization. You may see one for “Outside Spending Group” and one for “Political Nonprofit.”

The “Outside Spending Group” page shows data reported to the Federal Election Commission and the IRS. There is a tab for Targeted Candidates – note that these are the candidates the group is trying to defeat, not support. Since many of these groups aren’t required to report their funding sources, the Outside Spending Group page will reflect that lack of data.

The Political Nonprofits page provides additional data that the Center for Responsive Politics (which runs OpenSecrets.org) has compiled from IRS records, on non-profits that aren’t required to report their donors. But since these groups are secretive by nature, OpenSecrets.org doesn’t have donor information on all of them.

Who paid for this political TV ad? What are the ad-spending trends in my state and in the U.S.?

See the Federal Communications Commission and Political Ad Sleuth.

TIP: The FCC site allows look-up by TV station. Search by station call sign, then click “Political Files” (6th from the left in the row of buttons), then year, then race type and finally candidate name, and you’ll see PDFs of ad orders you can view. All U.S. TV stations are now required to put this information online. Radio stations are exempt for now and can keep such “public files” on paper at their offices.

Political Ad Sleuth lets you search ad buys across the U.S.. You can put in the name of an advertiser, for example, and see their ads grouped by state. The information on the site is not complete, but developers the Sunlight Foundation and Free Press call it “the most comprehensive data available.”

Can I get a report card on this senator/representative for the 113th Congress?

See Govtrack.us.

TIP: This website will tell you how often members of Congress missed votes, how often they joined in bipartisan bills, and whether their bills became law.

From the link above, scroll down to the member of Congress you are looking for, or search by name.

Can I see a list of all votes taken in the 113th Congress?

See OpenCongress.org

TIP: This page provides a reverse chronological list of all votes taken in the 113th Congress. Click on one of the bills or resolutions to see a breakdown by party and a list of each member’s votes.

Is crime on the rise in the U.S.? In my state and city?

See Crime in the U.S. and the Uniform Crime Reporting Statistics .

TIP:  The Uniform Crime Reporting (UCR) database is often cited for violent crime and property crime statistics, but has significant limitations (see list below). The database includes national and state crime estimates dating back to 1960, and city and county crime counts dating back to 1985. Covered crimes are criminal homicide, forcible rape, robbery, aggravated assault, burglary, larceny-theft, motor vehicle theft and arson.

You can browse the data from the Crime in the U.S. homepage , but it’s often easier to find what you need using this UCR tool. This will give you state-by-state and national estimates, by number of crimes and as crime rates, for each of the seven measured crimes.

According to these stats, the violent crime rate has fallen more or less steadily since 1991, from 758.2 per 100,000 population, to 386.9 – its lowest level since 1969.

The property crime rate reached its highest levels in 1980 (5,353.3 per 100,000) with another spike in the late 1980s and early 1990s, and since then has fallen to 2,859.2.

The UCR has several significant limitations, however.

• Participation is voluntary. The FBI says that participating law enforcement agencies in 2007 represented just shy of 95 percent of the US population. 
• The UCR employs a “hierarchy rule,” where in a multi-offense incident, only the most serious offense is reported. (There are exceptions for justifiable homicide, motor vehicle theft and arson.)
• The FBI also warns that UCR data should not be used to compare cities or counties, because the statistics do not take into account the many variables affecting crime reporting and crime rates, so such rankings present an overly simplistic picture.
• Reporting agencies can differ in how they classify crimes – and may even misreport crime. In reporting for the Dallas Morning News, Tanya Eiserer and Steve Thompson found that police had made it harder for store owners to report thefts, and this change accounted for much of the city’s touted crime drop. A Journal Sentinel investigation found more than 500 incidents misreported to the FBI as minor assaults.
• In the UCR’s Summary reporting system, rape is defined as carnal knowledge of a female, forcibly and against her will, so rape against male victims cannot be reported. These must be reported as assaults or other sex offenses. Rape against males can be reported under the FBI’s National Incident-Based Reporting System, but that system only covers 25 percent of the US population.
• UCR data includes reported crimes only. For unreported crimes, reporters may wish to see the Bureau of Justice’s, though that too has its issues. (See below: How common is rape on college campuses?)
• For more help, see this Investigative Reporters and Editors guidance on using FBI crime data.

Are mass shootings becoming more common?  

See pieces in New York magazine and Mother Jones. 

TIP: The short answer is: it depends how you define “mass shooting,” and it depends how you slice the data.

Northeastern University professor James Alan Fox, who is frequently interviewed by the media, contends that there has been no increase in shooting incidents  with four or more fatalities between 1976 and 2012 — although the data he provided to New York Magazine does show an upward trend in number of victims. In a 2013 paper, Fox and fellow criminologist Monica J. DeLateur found “the facts clearly say that there has been no increase in mass shootings and certainly no epidemic,” Slate reports.

The FBI has established four victims as the threshold for a “mass murder.”  Separately, Public Law 112-265 defined “mass killing” as three or more killings in a single incident, for the purposes of the legislation. The bill, signed in 2013, authorizes assistance to states investigating public violent crimes.

But more recently, Harvard and Northeastern University researchers Amy P. Cohen, Deborah Azrael and Matthew Miller published research showing that mass shootings have increased. This research, based on mass shootings data collected by Mother Jones, also looks at shootings in which four or more people were murdered. But it employs a very different methodology from Fox’s: Cohen et al restricted their study to shootings in public place, in which the shooter and victims were generally unrelated and unknown to each other. This is in contrast to Fox’s research, which includes all motives, such as domestic violence and gang violence.

Which definition is better? As Vox points out, “Declaring one or the other definition the ‘right’ one is too pat; each is right for the thing it’s right for.”

Besides the varying methods, reporters also need to be aware of potential flaws in government data. A USA Today investigation of 156 incidents found FBI data had an accuracy rate of only 61 percent, with several mass killings reported as unrelated single homicides, and at least a dozen crimes mischaracterized as mass killings.

With thanks to Mother Jones (further articles here and here) and On the Media. 

How common is rape on college campuses?

See this page from the Justice Department’s National Institute of Justice (NIJ).

TIP: Several studies indicate that between 18 and 20 percent of female students experience rape or other sexual assault during their college years, NIJ says. An oft-quoted statistic that one in four women will be raped in college is “not supported by the scientific evidence,” the institute says.

The Campus Sexual Assault Survey is often cited as the basis for the statistic that one in five college women have survived rape or attempted rape. But there are several issues with this 2007 study, conducted for the Justice Department’s National Institute of Justice by researchers at RTI International, the University of North Carolina and University of Cincinnati. The researchers surveyed women and men at just two institutions, hardly a representative sample. Also, surveys asked if the students had experienced rape or attempted rape at any point in their lives — not just at college itself.

Also note that this paper does not itself give the one-in-five stat. That number is, however, supported by a 2009 paper from the same team, analyzing college seniors’ assault rates over the duration of their college years.

In December 2014, the Bureau of Justice Statistics (BJS) published Rape And Sexual Assault Among College-Age Females, 1995-2013, based on data from the bureau’s National Crime Victimization Survey (NCVS). The BJS paper found that college-age females experience rape and sexual assault at a rate of 6.1 per 1,000 – much lower than previous estimates.

But a National Research Council report, commissioned by the BJS, found it likely that the NCVS undercounted rape and assault, with survey errors including ambiguous wording and a lack of privacy for respondents.

Interestingly, the BJS paper found that the rate of rape and sexual assault was 1.2 times higher for non-students than for students – but rape and sexual assault of non-students was also more likely to be reported to police.

Further data is available from the National Intimate Partner and Sexual Violence Survey but this survey does not identify student status among its respondents.

With thanks to The Washington Post’s Fact Checker  and Inside Higher Ed.

How many people do police officers kill every year? How often are the officers charged and convicted?

See this analysis by FiveThirtyEight.

TIP: The short answer is: We don’t know for sure. As FiveThirtyEight reports, the government does not keep data on unjustifiable homicides by police.

The FBI’s Supplementary Homicides Report estimates the number of justifiable police homicides as about 400 per year.

The Centers for Disease Control and Prevention’s National Vital Statistics System reports “homicides by legal intervention” and the Bureau of Justice Statistics estimates “arrest-related homicides.” Both yield numbers of about 400 per year. But both the FBI and the CDC undercount the number of people killed by police, for many reasons, FiveThirtyEight and PolitiFact say. Some reporting is voluntary, and the people recording deaths often don’t even know that officers were involved.

“The consensus among experts is that the numbers are too low, but it’s unclear by how much,” PolitiFact says.

The CDC has attempted to provide data on how many police homicides are ultimately determined to be justifiable, in the National Violent Death Reporting System. This data says that 95 percent of police deaths were reported justifiable. But the data is far from representative, because only 18 states participate, the Sunlight Foundation reported in August 2014. (The next month, the CDC announced it had awarded $7.5 million to expand NVDRS to 32 states.)

As for the frequency with which officers are charged, Philip Stinson of Bowling Green State University found a total of 81 cases from 2005 to 2011 in which at least one officer was charged with murder, non-negligent manslaughter or negligent manslaughter.

But with the data on number of killings by police so uncertain, it’s close to impossible to calculate what percentage of these cases result in a police officer being charged or convicted.

With thanks to Journalist’s Resource.

What days have the most drunk driving deaths?

The U.S. Department of Transportation provides this data through its Fatality Analysis Reporting System (FARS). We have the latest figures here.

TIP: FARS provides data on the highest blood alcohol concentration among drivers or motorcycle riders involved in fatal crashes, using a threshold of .08 grams per deciliter to determine what percent of crashes had alcohol-impaired drivers. The data shows that among six holiday periods (varying in length), the top holiday for drunken driving varies from year to year, with the percentage of alcohol impairment falling between 30 and 49 percent for all holidays in all reported years.

In 2013 the highest percentage of alcohol impairment was for New Year’s Day (44 percent), followed by the Fourth of July (39 percent), Labor Day and Memorial Day (both 38 percent), Christmas (37 percent) and Thanksgiving (33 percent). The highest number of fatalities in which a driver was alcohol-impaired occurred was 199, over the Fourth of July period.

The Insurance Institute for Highway Safety uses FARS data to calculate the worst calendar days of the year for drunk driving fatalities. IIHS found that on this basis, from 2008 to 2012, January 1 was by far the worst date, with 54 percent of crash deaths related to alcohol, and July 4 came next with 42 percent. Many of the remaining dates are not obviously related to any holiday. IIRS also calculates the worst months, and by this reckoning April, May, June and July tie at 32 percent.

 Is man-made climate change happening? Is there a scientific consensus?  

See “Quantifying the consensus on anthropogenic global warming in the scientific literature.”  

TIP: “Quantifying the consensus on anthropogenic global warming in the scientific literature”, a 2013 paper by Cook, Nuccitelli, et al, is the most well-known review of the subject. The researchers studied 12,000 peer-reviewed abstracts published between 1991 and 2011, and found that of those taking a position, over 97 percent agreed that climate change is man-made. Cook et al also contacted over 8,500 of the authors and asked them to classify their own papers (not just the abstracts). This again found that of those papers taking a position, 97 percent agreed that humans are causing global warming.

A 2009 paper by William R.L. Anderegg et al  (including one researcher from the William and Flora Hewlett Foundation, which also helps fund the API’s Fact-Checking Project), found that between 97 and 98 percent of climate researchers most actively publishing in the field support the tenets of man-made climate change as outlined by the Intergovernmental Panel on Climate Change. The paper also found that those who believe in man-made climate change have substantially more climate expertise and scientific prominence than researchers who do not. 

Another study, Examining the Scientific Consensus on Climate Change, by Peter T. Doran and Maggie Kendall Zimmerman of the University of Illinois, found that out of more than 3,000 earth scientists, 82 percent thought human activity is a significant contributing factor in changing mean global temperatures. For climate change specialists, who numbered 79 individuals, 97.4 percent agreed.

In a 2004 study, “The Scientific Consensus on Climate Change,” Naomi Oreskes (then of the University of California-San Diego) analyzed 928 abstracts with the keywords “climate change,” published from 1993 to 2003. She found that zero rejected the consensus position. (Oreskes found that 75 percent backed the consensus position explicitly or implicitly, and 25 percent dealt with methods or paleoclimate, taking no stance on current man-made climate change.) 

In addition, the scientific consensus is backed by organizations including the American Association for the Advancement of Science, the American Meteorological Society, and national science academies around the world. The UN’s Intergovernmental Panel on Climate Change confirmed its support for this position in 2014 with the publication of its Fifth Assessment Report, summarizing the current scientific knowledge on climate change. 

With thanks to PolitiFact. 

Is it true that global warming stopped for the past few decades?  

See Skeptical Science, UK Met Office.  

TIP: According to some analyses, judging by surface temperature alone, 1998 was the hottest year on record. However, this doesn’t mean that global warming has stopped – far from it. For one thing, 1998 was an outlier year. A study by Kevin Cowtan and Robert G. Way looked at global surface temperatures around the world, and found that temperatures increased 0.11 to 0.12°C per decade from 1997 to 2012, Skeptical Science says. Think of global warming as a long flight of stairs – we may have temporarily hit a landing, but the overall trend is still upwards. 

Even more importantly, less than 3 percent of global warming heat goes into increasing surface air temperature, while 90 percent goes into warming the oceans. The total heat content of our oceans, land, ice and atmosphere has risen decade by decade since the 1970s. Rises in ocean temperature are important because the oceans and atmosphere can exchange heat. In 1998, El Nino caused an unusually large heat transfer from the ocean to the atmosphere, leading to an above-average surface temperatures. 

The bottom line on plateauing surface temperatures is, even if they have occurred, they do not invalidate climate change models or alter the consensus scientific views that the Earth risks substantial warming by the end of the century. 

With thanks to PolitiFact. 

Are scientists manipulating climate data?  

See reports by the UK House of Commons Science and Technology Committee, U.S. Environmental Protection Agency (supporting documents here); U.S. Department of Commerce, National Science Foundation, University of East Anglia (independent panels led by Lord Oxburgh and Sir Muir Russell, and Penn State University in February 2010 and in July 2010. 

TIP: The idea that scientists are fabricating climate data comes mostly from “Climategate,” the 2009 hack of emails from the Climatic Research Unit at the University of East Anglia, in the UK. Climate change deniers have pointed to certain language in the emails, such as references to data “tricks” and “hiding the decline,” as evidence that researchers were manipulating data to achieve the results they wanted. Subsequent investigations by the UK and US governments, as well as by academic institutions, have cleared the scientists of any serious wrong-doing. 

With thanks to FactCheck.org.

Can we blame climate change for this particular weather event?  

See UN Intergovernmental Panel on Climate Change’s fifth assessment report, Working Group 1 and Explaining Extreme Events of 2011 from a Climate Perspective. 

TIP: There’s a big difference between climate and weather. Just because climate change is happening doesn’t mean every winter is going to be a warm one, everywhere on earth – variations from place to place and year to year are consistent with a long-term warming trend. 

Climate scientists do predict that the frequency of extreme weather events will increase. According to the UN Intergovernmental Panel on Climate Change’s fifth assessment report (Working Group 1, summary for policy makers), “Extreme precipitation events over most of the mid-latitude land masses and over wet tropical regions will very likely become more intense and more frequent by the end of this century, as global mean surface temperature increases.” The report also says it is “likely” (expressing “medium” confidence) that the intensity and duration of droughts will increase on a regional to global scale; and it is “more likely than not” that intense tropical cyclone activity will increase in the western North Pacific and North Atlantic. 

Because of the probabilistic nature of weather, and the many factors that determine the frequency and severity of weather events, it is impossible to say that a single event was “caused by” global warming in a simplistic sense. But, it is possible to make claims about the probability of a given event with and without climate change, according to researchers from the National Oceanic and Atmospheric Administration and the UK Met Office (citing this informational video). 

“One analogy of the effects of climate change on extreme weather is with a baseball player (or to choose another sport, a cricketer) who starts taking steroids and afterwards hits on average 20 percent more home runs (or sixes) in a season than he did before (Meehl 2012),” the researchers write. “For any one of his home runs (sixes) during the years the player was taking steroids, you would not know for sure whether it was caused by steroids or not. But you might be able to attribute his increased number to the steroids. And given that steroids have resulted in a 20 percent increased chance that any particular swing of the player’s bat results in a home run (or a six), you would be able to make an attribution statement that, all other things being equal, steroid use had increased the probability of that particular occurrence by 20 percent.” 

With thanks to Journalist’s Resource. 

What’s the trend in U.S.greenhouse gas emissions?  

See U.S. Energy-Related Carbon Dioxide Emissions, 2013.  

TIP: The trend very much depends on the time frame you choose. According to U.S. Energy-Related Carbon Dioxide Emissions, in 2013, US energy-related carbon dioxide emissions rose 2.5 percent from the year before, which was the fourth-highest yearly increase since 1990. But from 2005 to 2013, emissions fell 10 percent. Data on this webpage comes from the Monthly Energy Review of September 2014, unless otherwise indicated. 

Looking over a longer scale, from 1990 to 2013, energy-related carbon dioxide emissions in the U.S. increased on average by about 0.4 percent per year, according to EIA’s Monthly Energy Review for December 2014. 

Which states are the biggest greenhouse gas emitters?  

See State-Level Energy-Related Carbon Dioxide Emissions, 2000-2011. 

Per capita, the biggest emitters are Wyoming, North Dakota, Alaska, West Virginia and Louisiana.

TIP: The states with the highest energy-related emissions in 2011 were Texas, California, Pennsylvania, Ohio and Florida. It should be noted that the EPA counts emissions at the point where electricity is generated, not where it is used. States’ emission rates are influenced by their size, types of business, generation mix, climate and population density. 

Who is this social media user?

Try Facebook Graph Search, LinkedIn, Twitter Advanced Search, Topsy and Webmii.

TIP: Start by looking at the person’s profile, their posts and their connections – both on the social media site in question, and on other sites (Twitter, Facebook, LinkedIn, etc). These allow you to search using criteria such as location, occupation, age, current and former employers, and so on.

Topsy allows you to search Twitter for people, links, tweets, photos and video. It’s a good tool to use if you want to search for tweets from a specific time period.

WebMii lets you search by name to find a variety of media: photos, Wikipedia biographies, videos, news (from Yahoo), blogs (via Google) and social media profiles. The “web visibility” score can help you identify fake profiles.

Did this person really tweet this? Could this Twitter or Facebook account be a hoax?

Check out WebMii.

TIP: Look at the person’s Twitter profile to see how long he or she has been tweeting. If the account was opened very recently, it may be a fake.

Other aspects of the Twitter or Facebook profile to check out include location, friends and followers, linked websites, posted pictures and videos, and previous tweets or status updates.

WebMii is another tool to try – its “visibility score” (out of 10) helps identify fake profiles.

A lot of celebrities’ and politicians’ accounts have blue verification check marks, both on Twitter and Facebook. To see if the check mark is real, hover over it with your mouse. If it’s real, the text “verified account” will pop up. This may not be enough to be truly certain, however, and you should use all the tools at your disposal to try and verify the account.

Keep in mind how easy it is to fake a tweet. If you see a screen capture of a tweet, a good start would be to go to the user’s Twitter page to look for the tweet itself. That’s only a first step, though. If it doesn’t appear on the Twitter page, that could mean the screen grab was a hoax (created by a service like LemmeTweetThatForYou) – or it could just mean the user deleted the tweet.

Another way to fake a Tweet is to write a fake retweet. For example, “Whoa! @WhiteHouse Obamacare was a big mistake.”

Could this Twitter user actually be a bot?

Try Bot or Not?, an initiative of the Truthy Project at Indiana University.

TIP: Enter a Twitter handle, and the site will analyze user data, retweets and mentions to give you a percent probability that the Twitter user is actually a piece of automated software.

What’s this person’s address and phone number? I only have a phone number – can I find the person’s name?

Facebook Graph Search, LinkedIn, Twitter Advanced Search, Pipl.com, Spokeo.com, SearchSystems.net, Rapportive, AnyWho, AllAreaCodes, GeoSocial Footprint, Numberway, Google Person Finder.

TIP: Social networks can sometimes make the person’s location clear, so they’re a place to start. (But be aware that locations also can be faked.)

Then move on to online directories and people search sites. Each of these sites has its flaws and limitations, so you’ll probably get the best results by using several in combination. Here’s a run-down of some of the most popular among journalists:

Pipl.com allows you to search for a person by name, username, email address or phone number, with the option of specifying location. Returns photos, age, possible locations and websites referring to the individual.

Spokeo.com can find people by name, username, email or phone. In our test run of the limited free offer, we did find some inaccuracies. For example, Spokeo.com claimed five different “current addresses” for an individual, rather than correctly classifying these as past addresses. It does let you enter a person’s name and gradually narrow down by state, metro area, city and street. More complete results cost $4-$5 a month on a 3-month or 6-month plan. This provides profiles showing gender, age, occupation, education, marital status, economic profile, photos and other details.

SearchSystems.net is a global directory of free public records, with a people search function. Can return address history, relatives, employment history, and educational history.

Rapportive is a plugin that will give you information on your Gmail contacts, including location, employer, photo and social media profiles.

AnyWho is a Yellow Pages-owned website that allows you to look people up by name, address or phone number.

AllAreaCodes lets you enter a phone number to search for the person’s name. Names will be returned free of charge for numbers listed in the phone book. For other numbers, you must pay a small fee.

GeoSocial Footprint extrapolates where a user has been from clues on Twitter: their profile, check-ins and GPS-enabled tweets, and from natural language searches. This doesn’t work for every user – it depends how careful the user is with his or her locational data.

Numberway is a directory of international yellow pages and white pages. Enter the country you’re searching for, and you’ll get a list of links to online phone books in that country. As you might expect, listings are more complete for some countries than for others.

Google Person Finder is a crowd-sourced missing persons registry that Google runs during disasters.

Who owns this internet domain? Who hosts it?

WhoIs, HostCabi

TIP: WhoIs can give you the company name, email, physical address and phone number associated with a website (although results do vary). You’ll also see when the domain was last registered. If it was very recently, that could be a sign of a hoax. (Note that you want https://who.is, not the similarly named whois.com or whois.net.)

HostCabi allows you to enter a domain to find out where a website is hosted.

Is this article or blog post original?

Check out Churnalism

TIP: Enter a URL or text of an article, and this Sunlight Foundation tool will compare it to a database of press releases.

Has someone already debunked this?

Check out Snopes, Emergent, Hoax Slayer, Antiviral and What Was Fake on The Internet This Week

TIP: These sites are all good starting points to see if the rumor or news item you’re considering has already been debunked. If you don’t see the item on these sites, of course, you’ll have to keep digging.

Also, note that Antiviral posts from April 2014 and before are available here.

Did the event reported in this tweet really happen?

TweetDeck, Geofeedia

TIP: You can start by doing some additional searching in Twitter.

Run a search on a Twitter platform like TweetDeck and if you like limit this to tweets with pictures and video. You can also use a location filter. This will show you if anyone other than your original source has witnessed and tweeted the same event.

It’s helpful to set up your Twitter app ahead of time with the verified accounts of agencies and officials relevant to your interests. That way, when news breaks, you can quickly search their accounts all at once. (Thanks to the Verification Handbook and Circa’s Anthony De Rosa for these tips.)

Geofeedia is a commercial service that allows you to monitor social media content by location in real time.

When was this picture taken? What kind of camera took it?

Jeffrey’s Exif Viewer, Findexif.com

TIP: A good place to start is to check the image’s EXIF data, which can tell you the date and time the photo was taken and modified, what kind of camera was used, camera setting (lens, aperture, exposure, flash, etc.), and, where available, GPS location.

Note that the URL you enter should be the URL for the image itself, not for the website containing the image. This can usually be found by right-clicking on the image in your browser.

These tools can have trouble with social media sites, however, which strip out most of the original metadata. Flickr seems to be an exception.

Has this picture been altered?

Check out FotoForensics, Image Edited?, Izitru, JPEGSnoop and Imgops.

TIP: FotoForensics helps you figure out if a photo has been Photoshopped, using error level analysis (ELA), which searches for differences in compression levels throughout the image.

If you don’t have prior experience in these techniques, it may take some time to get up to speed, but the site’s tutorials are a good starting point.

Image Edited tries to do some of the thinking for you, giving “yes,” “probably,” or “no” answers to the question “Is this image edited?” In some cases it will tell you the editing program used and the date and time the image was modified. But to tell the extent of changes and the types of changes made, you’ll need a more detailed analysis tool such as FotoForensics.

Izitru works in a similar way to Image Edited, although Izitru’s classification system is based on assigning a “trust rating.” If the file appears to have been edited, it can’t earn the highest trust rating. Those images that do earn high trust ratings get hosted on the Izitru site for others’ reference.

JPEGSnoop is a free Windows-only program that reports image metadata and can also help identify if an image has been edited. It can handle a variety of file types, not just JPEGs.

Imgops is a free website that links to a variety of image verification tools. Upload your image and from there you can see reverse image searches on Google, TinEye and other sites. You can also view hidden data, optimize the photo and analyze error levels (to judge whether the photo has been altered).

Does this image show what it purports to show? Has this photo appeared anywhere else online?

Check out Snopes, TinEye, Google Reverse Image Search, Fast Image Research and Imgops.

TIP: When you’re trying to verify a photo, it’s important to see if it’s appeared anywhere online. Misattributed photos are perhaps the most common type of photo hoax. You might also find that the photo you’re investigating is a Photoshopped fake, with pieces sewn together from other, earlier photos.

You also want to find the original so you can do more verification work, such as contacting the photographer. Usually, the image with the highest resolution (or largest size) is the closest to the original.

First, try a Google image search with “site:Snopes.com,” to make sure it isn’t something Snopes has already debunked. Also try searching Snopes for relevant keywords.

Next, try a reverse image search using TinEye and Google Reverse Image Search. These tools allow you to find other places the photo has appeared online. To use them, upload an image or enter its URL. Note that the URL you enter should be the URL for the image itself, not for the website containing the image. This can usually be found by right-clicking on the image in your browser.

The two tools are likely to return different results, so it’s helpful to try your search on both.

A caveat about TinEye: the URLs it returns are not always exact. You may be directed to a top-level domain URL and not find the image there. If you’re having trouble finding the image on the one of the URLs in the results, try doing a domain-restricted Google Image search for it (search for your term and add site:x, replacing x with the target URL.)

To use Google Reverse Image Search, from the Google Image page linked above, click the camera icon. Google has a tutorial on using the search.

Fast Image Research is a Firefox and Chrome extension that speeds up your reverse image search by retrieving results from Google Images and TinEye simultaneously.

Imgops is a free website that links to a variety of image verification tools. Upload your image and from there you can see reverse image searches on Google, TinEye and other sites. You can also view hidden data, optimize the photo and analyze error levels (to judge whether the photo has been altered).

When was this video taken? Is it original?

See this YouTube data viewer.

TIP: This tool from Amnesty International’s Citizen Evidence Lab extracts hidden data from videos on YouTube.

The tool can tell you exact upload time and help find earlier versions of the same video using a reverse image search. If you have several copies of the same video from the same date, the tool can help determine which is original.

Where was this image or video shot?

Check out Google Maps, Google Earth, Bing Maps, Wikimapia, Open Street Map, Panoramio, Flickr, GeoNames and Wolfram Alpha.

TIP: These tools are invaluable for figuring out if the claimed location of a video or photo actually looks the way it’s supposed to.

Think about where the cameraman must have been standing to get the shot. A geolocation search could reveal that such a vantage point is impossible. Look not just at geography and buildings, but signs and license plates. If you need to translate a sign, try using the optical character recognition tool Free OCR to extract text from the image, and then run that text through Google Translate.

Many of these services also feature user-generated images, which can help you see features from a different angle. Of course, as with anything crowdsourced, you can expect some errors and inconsistencies.

Google Maps and Google Earth have a lot of similarities. They both feature satellite and street-view imagery, which can be very helpful. Google Maps also offers historical street views. Google Earth has its own powerful features that include historical aerial imagery and easy viewpoint rotation. (Thanks to Citizen Evidence Lab and the Google Earth Blog for their explainers on these tools.)

Wikimapia can often give you greater geographic detail and offers a good search-by-category function (for landmarks such as mosques and hospitals).

OpenStreetMap is a crowd-sourced map site. Aid groups and citizens can use it in a disaster to map out where damage has occurred – something too recent to be reflected in Google Maps.

Panoramio lets you search for a location to find geolocated, crowdsourced images overlaid on a Google Maps layer.

Flickr also lets you find geolocated photos, searching by keyword and city, state, country, etc.

GeoNames is a geographical database of over 8 million place names – not just towns and street names but schools, hospitals, geographical features and much more. The website can list results or display them on a map.

Wolfram Alpha is an automated tool that uses a knowledge base to answer users’ questions. In Wolfram Alpha, you can find out what the weather was like at the day and time the picture was shot, and if the position of shadows makes sense. To use it, just type a question in natural language, such as “What was the weather in Dallas on October 10, 2014?” (Thanks to Craig Silverman’s Poynter News University course, Getting It Right: Accuracy and Verification in the Digital Age, for the guidance on Wolfram Alpha.)

It’s hard to make out certain features of this video. How can I slow it down, brighten it, etc.?

Try Tracker and VLC Media Player.

Is the politician in this video lying?

Try Truth Teller.

TIP: This Washington Post initiative checks the claims that politicians make in videos against a database drawing on PolitiFact, FactCheck.org and the paper’s own Fact Checker. Truth Teller then flags true and false statements. You won’t necessarily find the video you’re looking for, but it’s worth searching for videos related to federal politicians here and for Texas politicians here (at the Post’s partner site, the Texas Tribune).