Contributions to the development of standardized guidelines for the (text-based) analysis of framing

facebooktwittergoogle_plusredditpinterestlinkedinmailby feather

In confronting certain limitations otherwise describing empirical research on social movement framing, various scholars have suggested ways to device methods that offer a potential for methodological standardization.

Never just criticize; criticize and provide possible solutions for whatever the critique is aimed at. I find that to be a quite nice rule of thumb, and I’ll attempt to go some way toward honoring it with the content presented here. In the previous entry, I outlined what some scholars have considered problems concerning empirical research on movement framing, but I did so, however, without describing the ideas that have been proposed in order to better the situation. My focus in this and some future posts will be a couple of those scholarly responses. For reasons already briefly stated I’m particularly interested in looking at methodological approaches concerned with the analysis of text, and ones that afford some degree of standardization across studies.

Hank Johnston (e.g., 1995, 2002, 2005), for one, has vehemently stressed the importance of dealing with said methodological issues. However, not only has he pointed to possible problems – he has also suggested potential solutions. At a general level, this part of Johnston’s work and the ideas presented therein are based on earlier proposals by Gerhards and Rucht (1992), and focus not least on the recreation of substantive contents of collective action frames proper. The analytical interest therefore essentially concerns the concrete ideas possessed by participants in a given social movement – data extracted from, for example, interview material (Johnston, 1995, 2005).

Now, where’s the potential for methodological standardization in the model? First of all, the features permitting this actually don’t really tie to the analytical approach itself, but rather to how the analytical results are presented. Thus, because the potential standardization afforded here doesn’t concern the creation of analytical results as such, in this case it might be appropriate to talk about presentation guidelines rather than what I’d label measurement guidelines.*Schema2

The presentation guidelines offered by Johnston basically suggest a systematic presentation of the extracted frame contents and their relations as elements in hierarchically organized graphical schemas. This kind of schematic presentation – supposedly reflecting the actual cognitive structure of frames – not least affords certain kinds of comparisons. In particular, and given the case-specific information represented, the possibility lies especially in the comparison of data acquired from one single movement at different points in time. Johnston moreover suggests adding page and line numbers to the elements present in the schemas, making somewhat transparent and verifiable the road taken by the analyst to recreate a given frame from its textual basis. By providing these kinds of guidelines, Johnston’s ideas, arguably, embody a quite moderate way of potentially fixating (i.e., standardizing) the methodological approach used to empirically investigate some of the framing perspective interests.

In later work (e.g., Johnston & Alimi, 2012), Johnston follows a quite similar path, although now drawing inspiration from not least story grammar analysis (see, e.g., Franzosi, 2010). In its usual form, story grammar analysis provides one way of systematically grounding global meaning dimensions in local linguistic patterns. In extremely simple terms, the approach builds on the assumption that the standard Subject-Verb-Object (SVO) structure reflects the prototypical linguistic expression of a (social) event, sequences of which constitute complete narratives. By coding the elements of a text in certain ways – with a starting point being the concrete categories of actors and actions found in those SVO structures – it becomes possible to (e.g., quantitatively) summarize the codes and represent the global narrative expressed. Arguably, this may also be used to get at the basic semantic structure of linguistically manifested protest events and social movement processes (e.g., Shapiro & Markoff, 1998; Franzosi, 1999).

Incorporating some of these insights into their model, Johnston and Alimi (2012: 7) argue that the elements in the SVO structure directly reflect the general narrative elements of collective action frames; “(…) the aggrieved <subject> challenging <verb> the offending <object>.” On that basis, the authors aim to summarize the frames of single movement processes and track the development of those over time. In concrete terms, however, and in contrast to regular story grammar analysis, they do so by interpretively looking at primary and secondary literature on those movements, and from that qualitatively build general representative SVO structures that they take to reflect collective action frames at different points in time. In other words, the analysis is not directly guided by the story grammar logic concerning textual elements at the clausal level, but rather uses the SVO structure to simply summarize and model the researchers’ findings. The approach thus very much resembles Johnston’s earlier work outlined above in that, for example, its potential for standardization still mainly relates to a set of guidelines describing the presentation of analytical results.

In sum, the two related strands of work by Johnston and colleagues described here embody some of the most developed ideas and suggestions for text-based methods aiming to improve in various ways on the lack of standardized guidelines in framing research. The next entry will provide an introduction to another ambitious method proposing a set of analytical guidelines for analyzing both framing tasks and frame components. The method builds on some of the insights and tools from just mentioned approaches and offers, among other things, actual measurement guidelines.


* I take these terms to describe a useful division for the present purposes, conveniently capturing two general potential underpinnings of methodological standardization. Note, however, that I do not suppose an actual independence between the two aspects, as it is likely that, in a given analytical method, they are usually related in form.


Franzosi, R. 2010, Quantitative Narrative Analysis, SAGE Publications, inc.

Franzosi, R. 1999, “The Return of the Actor: Networks of Interactions Among Social Actors During Periods of High Mobilization (Italy, 1919-1922)”, Mobilization, vol. 4, no. 2, pp. 131.

Gerhards, J. & Rucht, D. 1992, “Mesomobilization: Organizing and Framing in Two Protest Campaigns in West Germany”, The American Journal of Sociology, vol. 98, no. 3, pp. 555.

Johnston, H. 2005, “Comparative Frame Analysis” in Frames of Protest: Social Movements and the Framing Perspective, eds. H. Johnston & J.A. Noakes, Bowman & Littlefield Publishers, inc., pp. 237.

Johnston, H. 2002, “Verification and Proof in Frame and Discourse Analysis” in Methods of Social Movement Research, eds. B. Klandermans & S. Staggenborg, University of Minnesota Press, pp. 62.

Johnston, H. 1995, “A Methodology for Frame Analysis: From Discourse to Cognitive Schemata” in Social Movements and Culture, eds. H. Johnston & B. Klandermans, University of Minnesota Press, pp. 217.

Johnston, H. & Alimi, E.Y. 2012, “Primary Frameworks, Keying and the Dynamics of Contentious Politics: The Islamization of the Chechen and Palestinian National Movements”, Political Studies, vol. 60, pp. 603.

Shapiro, G. & Markoff, J. 1998, Revolutionary Demands: A Content Analysis of the Cahiers de Doléances of 1789, Stanford University Press.

2 comments to Contributions to the development of standardized guidelines for the (text-based) analysis of framing

  • Brian Diedrich

    I am currently an undergraduate and I am starting a sociology graduate program in January. Please forgive me if my questions and responses are over simplistic.

    Does not the standardization of methodological analysis to SVO contexts merely reduce the movement to a mere Marxist context thus detracting from the aesthetic (cultural and ideological) origins of the movement, which in the context of Weber are the most important factors?

    The civil rights movement for example can be demarcated into several sub-categories (Chicano movement, African American movement, women’s suffrage) all with unique motivations and origins. Not all of these origins can be explained through a SVO standardized presentation of data (as I understand it). The general idea can be explained in the standardized SVO method but the explanation of why Hispanics would not unite with African Americans and why women’s suffrage groups did not welcome minorities with open arms cannot. They did not directly conflict with each other. They merely had different ideological origins and contextual motivations which influenced the actions they took and the stratus with which they aligned. For example; Hispanics created the other white strategy and sought further definitions of racial boundaries, while A.A.’s sought to dismantle the system (Wilson. S. H. 2003). It does not seem to me that the SVO presentation of data can adequately account for this difference because Hispanics were definitely aggrieved by the offences of inequality however, chose different methods of appeasing those offenses, even when they had significant precedents to realize that their strategy was “counterproductive” (Wilson. S. H. 2003 p. 148).

    I don’t understand how a standardized SVO presentation of an analysis of text (laws) can describe why civil rights movements did not unite as a collective with a common cause; the oppressed vs. the legal system. My question is, How does the SVO method account for these unique aesthetic and efficacious differences of social movements through a standardized approach?

  • Lockert

    Hi Brian

    Thanks for your comment – the issue you’re raising here is very interesting.

    First of all, I have to admit that I’m generally quite fascinated by neat and orderly arrangements – something that would, for example, characterize a rigid method of text-based frame analysis. That’s probably one of the reasons I’m focusing on exactly that in the above post, and perhaps also the reason I could be blamed for doing so in a rather uncritical way.

    Still, I – and, I suppose, most other people dealing with the topic – would agree with your general critique (supposing I correctly understand the specifics of your comment). The development of not least such more or less objectively defined analytical guidelines is without a doubt an extremely reductionist enterprise relative to the “real-world” phenomenon such analysis could possibly be expected to say something about. There’s simply a lot of interesting and relevant dimensions that aren’t readily extractable from a movement’s text material. At least this is the case when dealing with methods that in their workings treat texts in rigidly predefined, and thus, by definition, inflexible ways.

    Although I would generally agree with you on at least having to be aware of those limitations, I wouldn’t necessarily consider them actual problems. The way I see it, the ambition in, for example, the Johnston/Alimi framework treated above, is not a complete modeling of the ideational dimensions of social movement processes, but an admittedly narrow one. developing not least this kind of method is necessarily a give-and-take situation. By improving on the “objectivity” and rigorousness of the method, I suppose you’ll often have to sacrifice some depth. I’d say as long as you’re explicit about it, you’re good to go. A more immediate problem with, for example, the kind of methods treated here are, arguably, whether the fixed presentation or measurement guidelines are actually potent at analyzing exactly what they are supposed to say something about. In other words, simply coming up with a way of creating comparable analytical results won’t ensure that those results adequately reflect the part of reality they are supposed to reflect (something related to the idea of “construct validity”). Not least when it comes to text analysis and interpretation I find this to be a pressing issue, and something I’ll also touch on in future entries.

    Finally, I’m interested in knowing whether, according to you, the SVO-based method could be at least a useful part of a more holistic methodological approach to summing up and analyzing social movement framing?


Leave a Reply




You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>