**7.4.5 Category # 5 – Unclassified data**

The majority of the data were classified in one or other of the categories identified above. However there was a small minority that did not fit any of them. These remained unclassified. Two examples are as follows.


Table 5. Unclassified data.

In light of the foregoing, Table 6 presents the frequency count of the codes generated from the classification of the meaning units.


Table 6. Frequency count of codes emerging from units of meaning.

To the extent that the writing of overviews and the selection and writing of details are two important tasks in the construction of an annotation, the evidence is that these two skills were well represented in the data set. Together they accounted for 69% of the meaning-units coded. However, it should also be noted that they were not used in relation to one another: one did not see the typical pattern of overview followed by supporting details in a given annotation. Instead both were used across different annotations.

In terms of summary writing for an annotation, the issue arises whether all relevant aspects of this skill were accounted for in the data set.

Data MP70 was recognized as a member of this category, in spite of the somewhat tentative nature of the verb. The decision to include it was based on the fact that other attributes of the assertion were evident and the strength of the second part of the verb phrase could be viewed as overshadowing the caution in 'seem'. Data MP108 was also acknowledged as a member of this category, notwithstanding the inclusion of 'the writer' as the apparent source. One wonders whether there was any evidence in the source material to support this

The majority of the data were classified in one or other of the categories identified above. However there was a small minority that did not fit any of them. These remained

MP73 This book excerpt is based upon findings derived from a series of surveys.

MP134 Sun Kyong Lee pointed out that media uses were based on different studies.

In light of the foregoing, Table 6 presents the frequency count of the codes generated from

**Code Number of units of meaning %** 

Overview 56 34% Superimposition 12 7% Detail 58 35% Assertion 19 12% Unclassified 18 11%

To the extent that the writing of overviews and the selection and writing of details are two important tasks in the construction of an annotation, the evidence is that these two skills were well represented in the data set. Together they accounted for 69% of the meaning-units coded. However, it should also be noted that they were not used in relation to one another: one did not see the typical pattern of overview followed by supporting details in a given

In terms of summary writing for an annotation, the issue arises whether all relevant aspects

'categorically' stated fact.

Table 5. Unclassified data.

**7.4.5 Category # 5 – Unclassified data** 

unclassified. Two examples are as follows.

**Data ID Unclassified meaning units** 

the classification of the meaning units.

MP93 He used two papers to scrutinize his research.

Table 6. Frequency count of codes emerging from units of meaning.

annotation. Instead both were used across different annotations.

of this skill were accounted for in the data set.
