1. specify the treatment targets (sounds that they missed) in measurable terms
2. prepare the stimulus items (whatever event, questions, prompts, modeling prompts, use objects, pictures, syllable shape of the word, choose vocabulary that they will know, consider culture, ethnicity, age, and langage level of the client; 20 stimulus items are recommeded but 10 may be acceptable)
3. prepare a recording sheet (how are you going to collect the data; what words to use, how to organize)
4. Administer the baseline trials (3 types):
-discrete trials (STRUCTURED opportunities to produce a given target; lists (the speech sound you're looking for))
-evoked trials (when the clinician arranges stimuli and provides a model of the target response for the child to imitate)
-modeled trials (when the clinician manipulates the stimuli and provides a model of the target resposne for the child to imitate) looking at typically developing children and norms:
-syllableness (for omitted vowels, diphthongs, vocalic/syllabic consonants resulting in productions limited to monosyllales; using two-syllable compound words, such as cowboy, and then three-syllable word combinations, such as cowboy hat)
-single consonants (prevocalic /p,b,m.w/; postvocalic voiceless stops; pre- and postvocalic consonants, such as pup, pop; intervocalic consonants such as apple)
-/s/ clusters (word initial, word final)
-liquids (word-initial /l/, word-initial /r/, word initial /kr/, /gr/-after the child readily produces singleton velars; word inital /l/ clusters-after child readily produces prevocalic /r/)
-anterior and posterior contrasts (word-final /k/, word-initial /k/, /g/ for "fronters"; occasional /h/; alveolars/labials if "backers"