Opinion Polls in International Perspective: The Case of West Germany

by guest contributor Sonja Ostrow

One can hardly open a newspaper without being inundated by graphs and charts offering up the latest poll numbers on presidential candidates. Almost as prominent are poll results covering attitudes toward everything from religion to home ownership. While public opinion research is typically thought of in conjunction with election cycles, it is more broadly integrated into the fabric of modern life. As Sarah Igo has observed, it has become one of the chief ways we know ourselves as citizens and as humans.

Still from Welt im Bild reel from August 28, 1953.

Still from Welt im Bild reel from August 28, 1953.

Igo was writing about the United States, but the trajectory of polling in other countries teaches different lessons about the collection and dissemination of knowledge. The history of public opinion research in West Germany in the first few decades after World War II makes clear how national polls have an outward-facing function: not only do they inform people (and their governments) about themselves, they have also provided a methodology for demonstrating national change to concerned foreign observers.

Rather than being conceived primarily as a way to forecast election results, empirical opinion research in postwar West Germany was prized as an innovative method for assessing denazification and democratization. Unlike in the United States, where historians like Igo have identified a persistent obsession with “the average,” in West Germany, public opinion polling was intended to illuminate the extremes of public opinion, the potential threats to a nascent democratic state. Especially in the 1950s, West German opinion research was as much about projecting an image outward into the rest of the “free world” as it was about developing insights that would make governing at home more responsive and effective.

In the wake of World War II, Allied political leaders and their social-scientific advisors were convinced that empirically-grounded survey research would help them to assess and, eventually, to shape the political cultures of former enemy lands. Survey units trailed the combat branches of Anglo-American forces in Italy in 1943 and in Germany in 1945, and began polling residents on issues including access to food, living conditions, radio habits and preferences, and political stances. While public opinion polling was already an entrenched part of the political culture in the United States and England, it was World War II that facilitated the spread of public opinion research worldwide.

The earliest German surveys were conducted under the auspices of Allied occupation forces, who only gradually convinced Germans that they were neither Communist spies nor agents of punitive denazification procedures. But native West German opinion research institutes quickly emerged in the western zones. Following in the rhetorical and methodological footsteps of Americans like George Gallup and Elmo Roper, these institutes—most notably the EMNID Institute in Bielefeld and the Institut für Demoskopie at Allensbach—loudly argued that their surveys, each of which unearthed the opinions of a statistically representative sample of roughly 2,000 Germans, would aid in the establishment of a democratic political culture by facilitating communication between government and governed and granting a voice to those who would otherwise have none. What is more, they claimed that their methods were inherently democratic, since polled subjects were selected based on their fulfillment of statistical criteria, and the opinion of each statistical citizen held equal weight.

These institutes analyzed millions of survey responses each year, producing chart- and number-filled reports about the hopes, fears, and perceived realities of West German citizens. The surveys conducted by the institutes are notable for their variety: no topic, from cosmetics-buying habits to anti-Semitism, was off-limits. “Trend questions” were asked annually to trace shifts in opinion over the years. One recurring EMNID poll asked respondents, “Do you have the impression that we can count ourselves among the society of western peoples, or in your opinion are we still enemies as we were in the past?” In 1954, EMNID emphasized the increase in the number of respondents agreeing that “we belong completely” (23% in 1954, up from only 8% in 1951) on the front page of its weekly newsletter, the EMNID-Informationen (Issue 7/54).

In some cases, research institutes invited survey participants to compare themselves explicitly with other national subjects. EMNID, for instance, asked West Germans on an annual basis throughout the 1950s, “Do you have the feeling that Americans see us today primarily as friends, as strangers, or as enemies from the past?” The percentage of those stating that they were seen by Americans as friends grew from 19% in 1951 to 49% in 1954, while the percentage of those declining to answer the question shrunk from 22% in 1951 to 10% in 1954 (EMNID-Informationen 7/54). Both figures likely represent a shift not only in attitudes toward the United States, but in a willingness to answer questions touching on matters of geopolitical significance in uncertain times. But the question construction itself is worth considering: why would EMNID ask people to see themselves through the eyes of another nation? In doing so, could the opinion research institute have helped propagate the idea that identity was only created through observation from the outside? Other historians have elaborated on the theory that national identity is brought to the fore at moments of encounter which bring one’s own unique background into sharp relief. Empirical opinion research helped to systematize and “scientize” such moments of encounter and comparison.

Elisabeth Noelle-Neumann, co-founder of the Institut für Demoskopie, Allensbach on the cover of Der Spiegel (October 1953).

Elisabeth Noelle-Neumann, co-founder of the Institut für Demoskopie, Allensbach on the cover of Der Spiegel (October 1953).

The Allensbach Institute, through its publications and the tireless work of its co-founder, Elisabeth Noelle-Neumann, fostered this notion more assertively than did EMNID. How else to understand the decision by the Institute to publish a compilation of its opinion research for an English-speaking audience? Noelle-Neumann and her husband introduced the 1967 publication of The Germans: Public Opinion Polls 1947-1966 with the observation that “This is not a portrayal of the Germans based on second or third hand reports; it is the nation’s own description of itself” (The Germans vii). The Neumanns described this self-assessment of the German people as a useful corrective to the assumptions of foreign observers. “The editors feel … that the most fruitful attribute of this publication is the fact that it disproves, or at least casts doubt on, stereotype judgments of a nation by its neighbors. The Germans, on account of their role in world politics over the past century, have at times been exposed to collective repudiation, more than any other nation, with the inevitable consequence that the entire population was identified with small ruling groups … only the self-portrayal of groups in the form of poll results can project a picture that is comparatively objective” (ix). The authors presented opinion research as the means by which Germans clarified their identities to themselves and to international observers. They also noted the public opinion research would be impossible in anything other than a democracy—thereby resting their case that there could be an lingering concerns about West Germany. The Germans was thus a multi-leveled vindication of the postwar West German public, and a clear attempt to sever that public from any continued association with National Socialism.

Sarah Igo, and, more recently, Jill Lepore, among others, have brought to life the biases and assumptions that often shaped the carrying out and reporting of polls. Yet in practice, historians frequently continue to refer to poll results as stable sources for understanding popular experiences of history. Of course, in many cases poll results do provide insights that would otherwise be completely unobtainable; yet they can never stand completely on their own.

The polling institutes that emerged in West Germany after 1945 were themselves quite aware of the malleability of polls. The Allensbach Institute led the charge in identifying, publicizing, and even exploiting biases contained within certain question formulations. Noelle-Neumann commented in a letter exchange with another researcher that the results for one survey on anti-Semitism in West Germany had to be understood in context: the questions in the survey had been crafted in order to evoke higher levels of anti-Semitism to gauge possibility rather than everyday attitudes – again, they were searching for the extremes rather than the average (Noelle-Neumann to Diedrich Osmer, 26 Jan. 1954, Korrespondenz mit Instituten, 1951- , Archive of the Institut für Sozialforschung, Frankfurt). Noelle-Neumann observed in a later address at a global congress of opinion researchers that question construction remained the crux of her work. Despite the attention paid to the accuracy of various sampling methods—the debate between those favoring random versus quota sampling raged throughout the 1950s—she argued that question construction that was more likely to prompt enormous shifts in responses.

However, for Noelle-Neumann and other opinion researchers, this element did not make polls any less “scientific.” And as Anja Kruke has shown, for the media outlets that were increasingly hungry for “news” of any sort, the subjective aspects of polls did not make them any less desirable.

In the second half of the twentieth century, there were also frequent attempts to develop comparative studies of European nations based on opinion polls. Such projects were made easier by the spread of the American Gallup Institute into Western Europe through partnerships with extant native institutes. For example, EMNID became Gallup’s West German affiliate in 1955, and in 1962 the institute carried out the polling in West Germany for a study on opinions about a European community, commissioned by the Press and Information Office of the European Community. European integration proceeded in tandem with the Europeanization of public opinion research. At the forefront was West Germany and West Germans, whose views on rearmament, anti-Semitism, and refugees, among other topics, were (and remain) a pressing concern for those within and beyond German borders.

Sonja Ostrow is a PhD Candidate at Vanderbilt University and a Review Editor at H-German. Her dissertation examines the use of empirical opinion research to measure and influence political change in Germany after World War II.


  1. Thank you, Sonja, for this fascinating post, raising so many suggestive issues! As a British historian, I was wondering about parallels to the famous Mass Observation project, which collected all kinds of data on ordinary Britons’ lives and their observation of major world events from the 1930s to the 1960s. Do you know whether that’s something of which your West German pollsters were aware, or whether they imagined any connections between it and their work? (Sorry if that’s a question outside your purview–I was just curious!)


    1. Emily, I’m so glad you found the post interesting! I have often wished that something like a Mass Observation project existed in West Germany – but alas, no. The pollsters I study were indeed aware of its existence in Great Britain, but to my knowledge nothing approaching that level of involvement in the lives of individual subjects/observers was undertaken in West Germany. I’m sure that part of the reason was the sensitivity to anything resembling the Gestapo or spying-like activity after World War II. This sensitivity was something the West German pollsters and the interviewers who worked for the Allies very early on had to try to counter with constant assurances of anonymity and references to the scientific nature of their work. Especially early on, in some ways their success depended on convincing West Germans that they were merely representatives of a particular category rather than individuals with unique experiences to share. I’m plan to situate West German opinion research methodologically in a broader international context in my introduction, so hopefully your question is one I will be able to explore further!


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