Step 1: Identify Causal Statements
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118
119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135
136 137 138 139 140 141 142 143 144 145 146
The first task is to identify the causal statements from the documents (e.g., interview
transcripts or annual reports) (Axelrod, 1976). This process involves identifying the
cause and effect phrases and the linkage between them. Causal statements are statements
that imply a cause-effect relationship. Some of the key words used in identifying explicit
causal statements are “if-then,” “because,” and “so” (Axelrod, 1976). In addition to
explicit causal statements, according to Axelrod (1976), there are also implicit relationships
found in causal statements. The phrase may not contain the traditional key words
used to identify causal statements, but the causality of the sentence is clear within the
context of the text. Some “key words” that have been used in identifying implicit causal
statements are “think,” “know,” “use,” and “believe”. For example, the sentence “If I
want to get beyond where I am today, then am I going to have to go outside of the
business?” could be coded as an explicit statement since it contains the words “if” and
“then.” Additionally, the sentence “I don’t think gender should be an issue, I would
promote whoever is smartest” can be coded as an implicit statement. The statements in
the form of concepts and cause-effect relationships are captured in the language of the
Term Definition
Causal Map A network of causal assertions (cause/link/effect) that can be
expressed in a matrix or diagram form.
Causal Statement A statement (phrase or sentence) that contains a casual assertion, most
generally of the form cause/link/effect.
Coding Scheme A dictionary of terms (concepts or constructs) and definitions of those
terms (concepts or constructs). The coding scheme is used to simplify
the causal statements and corresponding maps.
Concept A word or phrase that captures the meaning or essence of a
participant’s phrase.
Construct A word or phrase that captures the meaning or essence of a group of
concepts.
Link The relationship or causal belief between two concepts (or constructs).
Raw Causal Map A causal map in which the concepts (constructs) are represented in the
language of the participant.
Raw Causal Statement A causal statement that is captured in the language of the participant.
Revealed Causal Map The assertions of causality the participant chooses to reveal to the
world.
Table 3. Causal mapping definitions
Figure 2. Revealed causal mapping process
participants (Narayanan & Fahey, 1990). Other examples of causal statements would
include:
1. Object-oriented development is easy because you think of everything as an object.
2. If I’ve got this object built up then I go back and actually try to write some of the
methods.
3. Once I have all of the information I need I think about what are the objects that will
be needed.
Depending on the type of data collection, IECM or TBCM, the coding process will differ.
If you are using TBCM, generally you are using public documents (e.g., annual reports),
which have been carefully crafted. The author of the document has (most likely) placed
emphasis on the sentence construction, grammar and intended meaning of each sentence.
In this context, the causal statements should be relatively clear and straightforward.
*The term “text” is used to represent both IECM transcripts and TBCM texts.
In contrast, if you are using IECM, the causal statements are often difficult to discern
(Kemmerer, Buche & Narayanan, 2001). In this case the participant sample plays a large
role in the ease (or difficulty) of coding. For example, if you are speaking to IS personnel
regarding their current project, they are usually quite articulate. In contrast, if your
research sample consists of IS students discussing a very technical topic, or respondents
discussing a sensitive topic (e.g., layoffs) the participants may have difficulty
expressing themselves. In addition, you will probably have several “starts and stops”
within the transcript. By this I mean an individual will begin to speak, stop and then restart
with the thought. This can present challenges when coding the transcript. In this case,
it is up to the researcher to discern the causal statement (if any) in the text. It is often
helpful to have an audio recording (if possible) to listen to the tone of the participant in
addition to the words.
Identification Rules
The guidelines, which have been adapted from Axelrod (1976), are provided to show
researchers how causal maps can be derived from texts. The coder must scrutinize the
text to record all cause-effect relationships within the text. The sentences or phrases that
are of interest to the coder are those that assert a causal relationship (A affects B). To
appropriately identify the causal statements the researcher needs a set of decision rules
to help guide the process. The rules are:
1. Some relationships are implicit in the phrase or sentence and a cause/effect
relationship cannot be found in the structure of the phrase. In this case the coder
should ask herself if the phrase implies a relationship between variables. If yes, then
the phrase should be coded as a causal statement (be careful not to insert bias into
coding implicit statements to create assertions).
2. It is important to maintain the original language of the participants as faithfully as
possible.
3. It is important to reflect the speaker’s statement in kind and number. If a speaker
states a relationship more than once, the coder should note the relationship each
time it is mentioned.
4. If a speaker agrees with an assertion made by someone else the coder should pay
close attention to the speaker’s wording. If the speaker is agreeing with the
assertion then it is recorded as a causal statement. If the speaker is merely
acknowledging the statement then it is not coded.
5. Assertions should be made within a sentence or two at most. Do not look for
assertions by linking paragraphs.
In addition to these basic guidelines, Wrightson (1976) has provided a listing of the
structural relationships that may be found within a text and how they should be coded.
See Appendix A for an adapted (and abbreviated) sample of these structures.
Table 4. Sample causal statements