Orientation Observation In-depth interviews Document analysis and semiology Conversation and discourse analysis Secondary Data Surveys Experiments Ethics Research outcomes



Social Research Glossary

About Researching the Real World



© Lee Harvey 2012–2020

Page updated 29 April, 2020

Citation reference: Harvey, L., 2012–2020, Researching the Real World, available at
All rights belong to author.


A Guide to Methodology


Jonathan Potter (2010, p. 25–31) explains the discursive analytic process in some detail using material on crying from his study of child protection telephone help lines.

Crying and Crying Receipts
The analysis of crying and crying receipts started by identifying all the calls in which crying appears and then transcribing the full sequence. This generated a corpus of 14 crying sequences—some were quite brief (just a few turns) while some went on for many minutes and many pages of transcript. One of the first research tasks was to build an extension of the Jeffersonian transcription scheme that would enable the different features of crying such as sobs, whispers, wet and dry sniffs and tremulous voice to be represented. This extension is described in detail in Hepburn (2004) along with a detailed account of the limitations of the contemporary psychological literature on crying for dealing with it as an interactional object. This fine grained description of crying is extremely time consuming—yet its value is that it provides a way of seeing how delicately the different activities in crying and crying recipiency are organized together. The architecture of this interaction is complex with each participant carefully monitoring the other and showing the result of this monitoring in the turns of talk that they take.
We can illustrate this with the following extract. Various characteristic elements of crying on the helpline are highlighted such as caller apologies (A), and CPO actions such as 'right-thing' descriptions (RT), 'take-your-times' (TYT) and what we have termed 'empathic receipts' (ER). Note also the characteristic layout of the kinds of materials used in discourse research with the extract number, the code for particular data source, the anonimized participant descriptions (CPO, Caller), and the line numbers that allow specific reference to parts of the extract.

Extract One: JK distraught dad
1. Caller: >.Hhih .hhihhh< 1
2. CPO: D'you want- d'y'wann'ave [a break for a ] moment.= <TYT
3. Caller: [ Hhuhh >.hihh<]
4. Caller: =>hhuhh hhuhh<
6. Caller: .shih
7. (0.3)
8. Caller: °°k(hh)ay°°
9. (1.8)
10. Caller: .shih >hhuh hhuh[h]<
11. CPO: [S]'very har:d when <ER
12. CPO: they're not there with you isn't it.= <ER
13. CPO: and [you're-] (.) you're tal:kin about it. <ER
14. Caller: [>.hhih<]
15. (0.8)
16. Caller: >.Hhuh .HHuh<
17. (2.1)
18. Caller: .shih
19. (0.2)
20. Caller: °.shih° (.) °°(Need) hhelp(h)°°
21. (2.5)
22. Caller: .HHhihh°hh°
23. (0.5)
24. Caller: HHhuhh >.hih .hih<
25. (0.7)
26. CPO: .Htk.hh Well you're doing what you can now to <RT actually 27. offer them protection and help though <RT
are:n't you. <RT
29. Caller: .Skuh (.) Huhhhh
30. (0.5)
31.Caller: °°I:'m not the(hehheh)re. Hh°°
32. (3.2)
33. Caller: .Shih
34. (0.4)
35. Caller: ~^I'm ^sorry.~ <A
36. CPO: An they als- well E-E-Eddie obviously al- thought <RT
37. you were the person to contact to get he:lp. <RT
38. Caller: Yeh. hh
39. CPO: F'which (.) ye know he turned to you: .hh <RT
40. (0.7)
41. Caller: .Hh[h°hhh° ]
42. CPO: [T'help 'im.]=didn't he. <RT
43. Caller: °°Yhhehhh°°
44. CPO: So 'e saw you as a person who could help in this <RT
45. situa[tion ] for him:. <RT
46. Caller: [.Shih]
47. (0.9)
48. Caller: .Hdihhhh hhhuhh
49. (0.2)
50. Caller: H^oh: s(h)orry. <A
51. (0.4)
52. CPO: .Htk s'^oka:y. kay.
53. (1.3)
54. Caller: .SKUH
55. (0.3)
56. CPO: It's distressing but it's also quite a shock <RE
57. isn't it I guess [(for you)] <ER
58. Caller: [.HHHHhih]hh HHHhuhhhh
59 (1.7)

60. Caller: ((swallows)) °Hhhoh dhear.°

The identification and characterization of the different elements of crying allows the analyst to see how they are consequential for the unfolding interaction.
First, note the way the take-your-time in line 2 is occasioned by the caller's sobbing that starts in line 1 and continues through to line 4. We can see how delicate the mutual attention in this interaction is as, despite the sobbing, the caller responds to the take-your-time with a whispered 'khhay' (line 8). One of the interesting features we found with interaction and crying is there is a considerable amount of 'live silence'—that is silence which the recipient would normally expect to be filled by specific sorts of turns (Hepburn & Potter, 2007). Ironically, perhaps, interaction work shows that silence is a major part of crying.
Second, note further on in the sequence the caller's tremulously voiced apology (line 35). We might think that the caller is apologising for the transgressive nature of sobbing over the phone to a stranger. However, a careful examination of where apologies appear in crying sequences suggests that they are more likely to be apologies for disruption of ongoing actions or failing to provide normatively expected contributions. That is, they are explicated better by understanding the consequences of crying for basic conversational organization. For example, in this case the CPO's assessment in 26–8 is followed by an extremely quiet and very disrupted second assessment on 31 (the normatively expected turn). The following delay from the CPO would allow the turn to be recycled, and the apology could be specifically apologising for the absence of this recycling.
Third, note the right-thing descriptions on 26–8 and through 36–45. These are constructed from information already provided by the caller, re-described to present him having done the right thing. Such descriptions seem designed to reassure the caller and move him out of the crying sequence. These descriptions are often accompanied by tag questions (e.g. 28 and 42), which may be designed to lead away from crying by encouraging agreement with the right-thing description.

Finally let us consider the topic of empathy. This has been a notion in areas of psychology since the early 1900s—but it tends to be conceptualized in terms of a cognitive image of one mind sharing the experiences of another. However, we have focused on how empathy is built, as a practice, in real time in live situations, where each party has available to them the talk of the other (Hepburn & Potter, 2007). We found empathic receipts to be built by combining two elements:
1. A formulation of the crying party's mental or psychological state.
2. A marker of the contingency, doubt or source of the mental state formulation.
The mental state formulations (e.g. it's distressing but it's also quite a shock) are typically clearly derivable from local features of the talk such as the amount of sobs and wet sniffs, combined with the caller's own prior formulations of their state. That is, the empathic moment is not a magical one of mind reading but a mundane and practical one involving responding to what is in the immediate talk and in doing so displaying close monitoring. The mental state formulation is combined with a second element which involves the recipient marking their formulation as limited ('I guess'), dependent on what is hearable, or using a tag question ('isn't it') to mark the speaker as the one with authority over the correctness of the formulation. In each of these ways the CPO defers to the caller who is crying as the party who has the right to define the nature of their own psychological state. What we have here is a procedural account of empathy grounded in the perspectives of the participants as displayed in their talk.
More generally, although emotion is often thought of as something that is beyond the purchase of discourse research (probably because of its early emphasis on people talking about things in open ended interviews), studies of this kind show the way that issues and actions which we understand as emotional can be tractable to interaction analysis (cf. Edwards, 1997, 1999). This is not surprising once we remind ourselves of the practical and communicative role that emotions play in social life (Planalp, 1999). Indeed, by carefully listening to these materials and carefully transcribing the interaction, the analysis starts to highlight precisely how the 'emotional' issues become live, are noticed, attended to, managed, and how both parties mutually coordinate in very fine ways in what seems initially like a highly chaotic strip of interaction.
In this case, the patterning may reflect one institutional setting and its goal orientations—much more work will be needed to develop an understanding of the complex patterning where speakers are familiars, where babies and young children are involved, where there is an immediate physical or psychological cause of the crying (Hepburn & Potter, forthcoming). More broadly, discursive work offers the possibility of understanding the various phenomena loosely glossed as emotion in terms of what they are doing and where they appear in peoples' lives.


Return to Discursive Psychology: Analysis (Section