## Interobserver Agreement Cooper

This technical report provides detailed information on the reasons for using a common computer computing program (Microsoft Excel®) to calculate different forms of interobserver agreement for continuous and discontinuous datasets. In addition, we offer a brief tutorial for using an Excel table to automatically calculate the traditional total number, partial match in intervals, exact tuning, trial test, interval interval, multiple interval, total duration and average duration of interobserver duration algorithms. We conclude with a discussion of how practitioners can integrate this tool into their clinical work. IOA intervals for intervals. In short, the interval interval method assesses the proportion of intervals in which both observers agreed to determine whether the target reaction occurred. Note that this implies agreement on attendance and lack of response. This is calculated by adding the total number of agreed intervals to the sum of agreed intervals and divided at regular intervals. Not surprisingly, this approach often leads to high convergence statistics. As Cooper et al. (2007) reports, this is especially true when partial interval recordings are used. In the examples in Figure 2, observers disagree on the first and seventh intervals, resulting in an interval agreement value of 71.4% (5/7).

Developers and users of slot analysis did not recommend that tolerance be reviewed with permanent measures (z.B. MacLean et al., 1985; Repp et al., 1989; Tapp – Wehby, 2000). Current data indicate that the same level of tolerance (±2 s) can be justified if duration is the measure of interest. A standard tolerance (for example. B ±2 s) for reporting accuracy and agreement for recording events or durations can help interpret the quality of the data evaluated using the period analysis algorithm. The unnecessary alternative, with several different tolerances for different measurements within the same data set, would probably make data analysis more difficult. Average duration pro-deposits IOA. If the number of calendars is high, it is important to limit data aggregation in order to identify possible variations in the permanent data of two observers. The average duration IOA algorithm per deposit achieves this by determining an IOA score for each timing, and then by deifing them by the total number of timings in which the two observers collected data. Note that this approach is similar to the approach described above of partial agreement at regular intervals. In the example of Figure 3, there were 99.7, 2.3, 69.2 and 92.7% approval levels for intervals 1 to 4, respectively. The average of these four levels of the agreement results in an average of 66% per event agreement – a much more conservative estimate than that of the statistics of the total duration of the IOA.

Observer accuracy, calculated as the conformity of observer behavioural records with test readings, was analyzed separately for the recording of observers on the various handheld computers (Psion and iPAQ).