The ten reasons why telephone surveys work
Since the time that telephone surveys were introduced as a viable way to sample opinions, there has been much debate about their effectiveness and how reliable they are. Over the decades, changing trends in the way that people live and communicate has made most of the criticisms of bias irrelevant to using a phone poll. Arguments that not everyone has one, or that only those of the higher socioeconomic status have one in their home have given way to a new technological age.
Many would argue that there is no better way to have the potential to reach anyone at any given time and in any given demographic than to call them on their telephones. Take a drive past any playground or any crowded doctor’s office, and you will see that if there is one thing that everyone not only has but never leaves home without, it is their phone. Comparing the past to the present, these are the ten reasons that phone surveys are the most reliable way to conduct opinion research.
1 Everyone has a phone
Of the millions of Americans across the nation, you would be hard pressed to find anyone who doesn’t own their own phone or have access to a household phone. With landlines giving way to the smart, or cell phone, the potential to reach anyone at any time is at an all-time high. Even though most telephone surveys use landline numbers only, there is no law against using cell numbers. With over 90% of all Americans having a telephone, there is no greater way to reach them.
2 They are more reliable than internet surveys
The new age of computer polling doesn’t hold up to the same scrutiny that telephone polls do. Telephone surveys have to be tested, and re-tested, to ensure that the survey used is measuring what you want it to. The same standards are not employed with computer polls. Internet polls may be less expensive, but they don’t have the same standards about taking all things into consideration. They are not as stable when considering things such as intention and behaviors of the people you are polling.
3 Telephone polls still have one to one communication with the subject
Unlike other forms of surveying people, such as the internet, when telephone polls are conducted they are done so using real people. That means that there is less chance that there will be a miscommunication between the poll and the person who is answering the poll. There is less probability that there will be confusion between what is being asked and the way that the person is reading it. If the person being polled isn’t clear, all they have to do is ask a question or say they are unclear. That yields better results and validity.
4 Even with the decline in answering, it is still the most reliable
There has been concern that many households have become exhausted by the continual “cold calling” and telemarketers. Those who apply to “Don’t Call” registries are still able to be contacted by government opinion polls since they are considered beneficial to the public and are, therefore, allowed by law. There are those who either refuse to answer after answering the phone, and then there are what is called the “silent refusal” participants. A silent refusal participant is a person who sees the number on their caller ID and chooses not to answer it. There has been questioning as to whether the drop in those willing to take phone surveys has led to a decrease in the representativeness of telephone polls. The 2004 Pew study concluded that even though there was a significant drop in those who are willing to participate in telephone surveys, “carefully conducted polls continue to obtain representative samples of the public and provide accurate data about the views and experiences of Americans…The decline in participation has not undermined the validity of most surveys conducted by reputable polling organizations.”
5 Studies show that landline users are comparable to cell phone users
Due to the shifts in technology, and the fact that many are opting to lose their landline and go wireless, many fear that it will result in a bias between those who are accessible through landlines and those who are cell phone users only. Many studies conducted have demonstrated that there is no inherent difference between those who use cell phones and those who use landlines. There is no observable bias resulting from the technological shift. When comparing the two groups, and their opinions, no disparities have been found.
6 There is the potential to add cell phone callers
If there does come a time when landline and cell phone users appear to be creating a bias, there is no law or regulation against using cell phone numbers for opinion polling. The reason they are currently not being used is because using cell phones for telephone surveys is much more expensive than using the standard landline. Although not illegal to use cell phones, it is illegal to operate an automated dialing system. That means that each number would necessarily have to be punched in by hand. That leads to more resources and man hours, which equate into time and expense.
7 Why zealots are not a bad thing
When we talk about voters and voting polls, there are those who insist that only those who have a strong opinion are likely to answer the phone and answer the survey, and that creates bias. When you rationally think about the argument, it becomes illogical to think that it is necessarily a bad thing, or in some way makes your responses less valid. In fact, if that is the case, it makes telephone polls even more valid. If you are likely to pick up a phone to answer a survey about how you are going to vote, then you are more likely to take the time to leave your home to vote. Therefore, if you are looking for those who are not just voicing an opinion but are going to take action, then you want that segment who are dedicated to having a voice. If they don’t want to have a voice, they are not likely to vote and maybe should not be considered in a poll measuring likely voting behavior, or predicting how a vote will go.
8 If prediction markets do better than traditional polling telephone polls are ripe for the change
There has been some argument that prediction markets are better than traditional polling. For instance, if you ask someone “who are you going to vote for in the 2016 election?” you are likely to get a different answer than if you ask “who do you think is going to win the 2016 election?”. There has been some evidence that prediction marketing is much more reliable to outcomes than the traditional surveys that are standard in polling. If things do change to polling people according to predictions rather than to what they are personally going to do, the telephone poll is going to be the best place to get people to answer. It is also a great way to ramp up a political race and get people excited about getting out and voting. If they make a prediction, they are likely to vote the way that they are predicting it will go, regardless of personal opinion.
9 Not everyone owns a computer
The biggest contender that phone surveys have in the field of polling is the internet poll. Although a much less expensive way to poll, the internet does not produce the same “randomness” as telephone surveys. Phone polls have a better random sample because not everyone has access to a computer. According to the Pew Research Center, as of 2014, 90% of all Americans had access to a phone, 58% of Americans have a smartphone, and only 42% of Americans own their own computer or tablet. Even those percentages can be misleading. Just because you have, or own a computer, does not mean that you have access to the internet. When you look at those percentages, it is fair to say that getting a random sample using a telephone is much more reliable.
10 Not all demographics go on the same internet sites
When you use an internet poll, you have to rely on the premise that there is an equal likelihood that everyone on the internet uses the same internet pages. We all know that that is not true. Therefore, when using an internet poll, you are going only to be polling those who use a particular website or webpage. That creates bias and reduces the likelihood that you are going to have a true random sample. When you reach people on the phone using telephone surveys, you are reaching out to them in a way where everyone has the same chance of being contacted, not only those who go to a specific web page.
Every polling method is going to have bias. It is virtually impossible to eradicate all bias completely from a survey of any kind. When dealing with humans, human nature, and technology, there are always going to be limitations on what you can and can’t do. The telephone survey has been, and continues to be, the best way to get a representative sample of what Americans are really thinking.