Tier Match

The Tier Match analysis compares aligned values from two tiers. By default, it compares each IPA Target word with its aligned IPA Actual word, reports exact matches, and calculates their character edit distance.

Use this analysis for a literal comparison of tier values. For phone-alignment accuracy, use a phone-based analysis instead.

Parameters

Table 1. Tier Match parameters
ParameterDefaultEffect
Report Title Tier Match Sets the title of the generated report.
Combine results for all selected participants Off Combines the selected participants before calculating comparison summaries.
Ignore truncated On Excludes a result when its aligned Tier 2 value is empty.
Tier 1 IPA Target Selects the primary tier and the value shown first in each comparison.
Tier 2 IPA Actual Selects the word-aligned tier compared with Tier 1.
Case sensitive On Treats uppercase and lowercase characters as different.
Diacritic Options Do not ignore diacritics For IPA tier values, optionally removes selected diacritics or all but selected diacritics before comparison.

The default Data Tiers query searches every non-empty IPA Target word and requires an aligned IPA Actual word. Standard word, tier, additional-tier-data, and participant filters can narrow or extend the results. See Common Query Parameters.

Figure 1. Tier Match Parameters
Tier Match Parameters

How tiers are compared

After applying the case and diacritic options, the analysis compares the complete text of Tier 1 and Tier 2. It also calculates Levenshtein distance: the smallest number of single-character insertions, deletions, and substitutions needed to transform one value into the other.

Word Count is the number of non-truncated aligned pairs. Match % and Mismatch % divide their respective counts by Word Count and multiply by 100. A session's Avg Distance is the sum of mismatch distances divided by all compared pairs, so matches contribute zero. The participant-level Avg Distance is the unweighted mean of the session averages that contain results.

Report output

The report contains a global Summary, a Readme, and one section per participant:

Table 2. Tier Match report tables
TableContents
Summary One row per participant with Word Count, Avg Distance, Matches, Match %, Mismatches, and Mismatch %. The participant section repeats these measures by session and includes age.
Matches and Mismatches Token listings with source-record information, the two compared tier values, character edit distance, and the comparison result.

This comparison is also included in Word Match.