Home > Uncategorized > REACTION 1: Why Can’t We All Just Get Along? A look at methods in political communication

REACTION 1: Why Can’t We All Just Get Along? A look at methods in political communication

A central concern of political communication scholars interested in political advertising is the inability of scholars to reach a consensus regarding the effects of advertisements on political knowledge, agenda-setting, and, in particularly, propensity to alter citizens’ views and votes in regards to candidates.  A common explanation for this inconsistency in research, as apparent by the readings for this week, is methodological.  According to Goldstein and Ridout (2004), and Daniel Stevens (2007), the lack of consensus in political advertising research is directly related to the disagreement among scholars on the appropriate method to use when determining the effects of political ads on citizen behavior and knowledge.  Furthermore, the different methods chosen by scholars of political communication have serious drawbacks that are difficult to overcome.

            Goldstein and Ridout discuss seven approaches to the study of political advertisements:  campaign spending, archival data, ad buys, tracking data, experimental manipulation, individual level recall, and Zaller’s political knowledge index.  Goldstein and Ridout assert that each of these methods, except for one, have serious drawbacks.  For example, individual-level recall, a method used extensively in the classic work by Charles Atkin and Gary Heald (1976), is riddled with problems.  First, people have an inherent problem accurately recalling when and how many times they have viewed a political advertisement.  Furthermore, ad recall and political behavior are endogeneous, or highly correlated to one another in such a way that it is difficult to determine what variable is the cause of the other.

            All the methods, except tracking data, contain similar problems.  Yet, although tracking data appears the best method in the Goldstein and Ridout piece, Daniel Stevens argue that tracking data has problems of its own.  While this method has been used more frequently in recent years, particularly with the advent of “ad detectors” in 100 major media markets pioneered by Campaign Media Analysis Group, Stevens believes that care must be taken in using such methods.  In his experimental study of undergraduate students and their television viewing habits, supplemented by ANES survey data, he finds that tracking data contains measurement error that can so alter scholars’ findings that opposing effects may be found.  Thus, according to Stevens, this new, creative measure is not the answer for political communication scholars.

            So, where are political advertising studies to go from here?  If no method is particularly useful, and no one can agree on the best method to use in studying the effects of political ads on knowledge, agenda, vote, and persuasive ability, what can scholars do?  First, I assert that the best way to avoid the problems associated with each method is to combine methods.  For example, Michael Franz and Travis Ridout (2007), in their study of political advertising’s power to persuade citizens, use two methods: Survey data from the Center for the Study of Elections and Democracy, and political tracking data from the Wisconsin Advertising Project. By using both methods to study the effects of political advertisements on persuasion, Franz and Ridout successfully overcome many of the problems associated with methodology. 

The key strength in superimposing one method over another lies in the tradeoffs a researchers confront when selecting a method of research.  For example, experimental methods have a high degree of internal validity because researchers can control the environment for which a study is conducted.  This prohibits other, unforeseen external variables from interfering with the observation that scholars are interested in observing.  However, experimental studies fail to achieve high rates of external validity; it is difficult to generalize results obtained from experiments to the real world unless scholars are successful in recreating, to the best of their ability, the real world.  Thus, combining methods, such as Franz and Ridout do, helps to overcome some of these tradeoffs.

However, I do not contend that the use of multiple methods in any given study is possible, or even reasonable, in all research projects.  Studies on political advertising can be incredibly time consuming, and using supplemental methods can sometimes be quite unrealistic.  Furthermore, researchers must take into consideration the research question they are addressing when choosing what method to employ, and some research questions are not amicable to multiple research methods.  In addition, in some cases, a research question can be adequately addressed using only one research method, making additional methodologies repetitive and unnecessary. 

Although I believe multi-method studies are best, if one method must be selected because of time or research question considerations, I suggest that scholars take advantage of the new tracking system when possible, or use an experimental design when a research question calls for it.  Despite Stevens criticisms, the new tracking data allows researchers the ability to take into account information, such as the frequency and time of day an ads were aired, into consideration when performing studies.  Furthermore, the data is so immense that the possibilities are seemly endless.  As for experiments, the ability to control independent and dependent variables is extremely beneficial to the study of political advertising, and offers much to determining the effect of ads on human behavior and thinking. One caveat is in order, though. Students of political communication would be wise to make their experimental settings as close to real life as possible.  Methods such as sending ads via digital mediums to a person’s home would greatly increase the validity of studies. 

In short, based on the readings for this week, I argue that multi-method studies are the most beneficial for the future of political advertising research.  However, when only one method is amicable to the research, I assert that experimental manipulation or tracking information is best.

Categories: Uncategorized
  1. No comments yet.
  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: