All the legitimate data and studies show that robo calls don’t work. Not now, not ever. So, I ask you robo call guys and gals, to show me the money.

Show us the data that you have that proves that robo calls do more than turn voters off from elections, candidates and democracy. I have 85,000 members that all detest robo calls and want to stop the madness.

But don’t believe me.

Here is a quote from a member of the National Political Do Not Contact Registry this week:

You can tell Eric Roberson, Democrat running for Pete Sessions's congressional seat in Dallas, TX, that I withheld my vote for him SOLELY because of the MANY robocalls he showered me with. Since I generally vote against the religious right wing, I would have voted against Sessions by voting for Roberson, but his half-dozen or so calls from 972-599-2485 ticked me off. You can tell him and other candidates that at least one voter thinks and acts this way.

OR, how about this one:

Since I am dying from cancer, I had my phone disconnected...solved the problem of answering unwanted call and the pain to answer them this last year! Nothing else seemed to totally work.

The facts are that politicians make robo calls because they can. Not because they work.

Here are some more facts, as reported by Todd Rogers at the Analyst Institute (

A 2005 study failed to show that robocalls generated mobilization results among youth voters. This study was conducted during the 2005 New Jersey Gubernatorial Election. 18,000 young voters were randomly selected to be robocalled with one of two GOTV messages, while the rest of the young voters in the election were left uncontacted. The first group’s message encouraged turnout and informed the voters of their polling location. The second group’s message was a generic turnout encouragement. Neither message was effective at increasing turnout over the uncontacted group (Nickerson, 2005).

Another study in 2001 in Seattle failed to find any mobilizing effect from robocalls when placed from a local election official. The local registrar of voters recorded a message sent to 10,000 voters the day before the November Municipal election reminding citizens to vote. No statistically significant increase in turnout resulted among those who received the robocalls relative to those who did not (Nickerson, 2002).

A 2006 study failed to find that robocalls increased turnout among Latino voters when the calls were placed from a Latino organization. The National Association of Latino Elected Officials (NALEO) conducted the study during the 2006 General Election in five California counties. Robocalls were delivered to 61,422 low propensity Latino voters. No effect on turnout was found among those who were called relative to those who were not called (Michelson, 2007).

A 2005 study failed to find that robocalls had an effect when using the voice of a Latino celebrity. NALEO conducted this study in California, New Mexico, and the city of Houston, TX, during the 2002 General Election. More than 250,000 phone numbers associated with Latino voters received two robocalls each. The calls were recorded in Spanish by a Spanish-language-television celebrity anchorwoman. The researchers found no statistically reliable increase in turnout among those who were assigned to receive the calls relative to those who were assigned to not receive the calls. The cost of the robocall campaign was $23,725 which means that the robocalls in this study resulted in a cost per vote of approximately $275 (Ramirez, 2005).

A similar 2004 study failed to find that robocalls had an effect when using the voice of an African American celebrity to call African American populations. In North Carolina and Missouri during the 2004 Presidential Election, precincts were divided into three groups. The groups were defined by precincts with high voter density and a large percentage of African American voters, precincts with a low voter density and a moderate percentage of African American voters, and precincts with a high voter density and a low percentage of African American voters. Individuals in these groupings were then randomly assigned to receive one of two robocall messages or to be left uncontacted. Both robodials were recorded by Vanessa Williams. The first message emphasized voting rights. The second emphasized the important issues of the election. In none of the groupings or message conditions was a measurable increase in voter turnout detected over the uncontacted control group (Green and Karlan, 2006).

Another study failed to demonstrate an effect from robocalls using political endorsements. This experiment was conducted during the Texas Republican Primary for State Supreme Court Justice. In this study, hundreds of thousands of Republican voters received a robocall message recorded by popular Republican Governor, Rick Perry. The message encouraged Republicans to turnout for the upcoming election, and to vote specifically for the endorsed nominee. Those who received the robocalls were not measurably more likely to vote than those who did not (Green and Gerber, 2008 p. 83).

So, guys, so me the money.

Shaun Dakin is CEO of the National Political Do-Not-Contact Registry.