EBP THERAPY ANALYSIS
Note: Scroll down near the end of the page to read the summary of the procedures.
C = Clinician
EBP = evidence-based practice
eta = partial eta squared
f = female
m = male
MLU = mean length of utterance
NA = not applicable
P = Patient or Participant
pmh = Patricia Hargrove, blog developer
SLP = speech–language pathologist
SOURCE: Thompson, W. F., Schellenberg, E. G., & Husain, G. (2004). Decoding speech prosody: Do music lessons help? Emotion, 4 (1), 46-64. DOI: 10.1037/1528-35188.8.131.52
REVIEWERS: Melissa Schachter (Molloy College; Rockville Centre, NY); pmh
DATE: May 1, 2016
ASSIGNED GRADE FOR OVERALL QUALITY OF EVIDENCE (for Experiment #3 only—See explanation in the next item for restricting the review to Experiment #3): Not Graded (The highest possible grade associated with this investigation was C+ because this was retrospective research. This investigation was not graded because its participants were typically developing. The lack of a Grade for Overall Quality of Evidence should not be construed to indicate that the procedure is without merit, rather it identifies the nature of the evidence supporting the effectiveness for children with disabilities.)
TAKE AWAY: This investigation comprised 3 experiments. The first 2 experiments are comparisons of adults with and without musical training. They are not reviewed here, although the reviews are available from the blog developer (Patricia Hargrove; pmh.) The 3rd Experiment was a retrospective examination which compared 4 groups of typically developing Canadian children who had been divided into 4 groups: keyboard instruction, singing instruction, drama instruction, and no treatment for 1 year. The results indicated that the keyboard and drama groups exhibited superior skills in distinguishing fearful and angry affective prosody.
- What type of evidence was identified?
- What was the type of evidence? Retrospective, Randomized Group Design with Controls?
- What was the level of support associated with the type of evidence?
Level = C+
- Group membership determination:
- If there was more than one group, were participants (Ps) randomly assigned to groups? Yes
- Was administration of intervention status concealed?
- from participants? No
- from clinicians No
- from data analyzers? Yes
- Were the groups adequately described? Unclear.
- How many Ps were involved in the study?
- total # of Ps: 43
- # of groups: 4
- List names of groups and the # of participants in each group:
– Keyboard Training = 10
– Singing Training = 11
– Drama Training = 9
– No Training = 13
- CONTROLLED CHARACTERISTICS
- cognitive skills: The Ps were typically developing (assumed)
- educational level of clients: first graders (during the intervention)
- DESCRIBED CHARACTERISTICS
- age: overall—
– 6 years (1st Grade, G1, during intervention in previous investigation)
– 7 years (between G1 and G2, during testing)
- gender: overall – 11m; 32f
- cognitive skills:
- Verbal IQ (no significant difference among groups pretest)
– Keyboard Training = mean = 106
– Singing Training = mean = 107
– Drama Training = mean = 108
– No Training = mean = 108
- Performance IQ (no significant difference among groups pretest)
– Keyboard Training = mean = 109
– Singing Training = mean = 117
– Drama Training = mean = 107
– No Training = mean = 104
- Overall IQ (no significant difference among groups pretest)
– Keyboard Training = mean = 108
– Singing Training = mean = 112
– Drama Training = mean = 107
– No Training = mean = 106
- educational achievement (no significant difference among groups pretest)
– Keyboard Training = mean = 109
– Singing Training = mean = 104
– Drama Training = mean = 106
– No Training = mean = 105
- behavioral skills (no significant difference among groups pretest)
– Keyboard Training = mean = 43
– Singing Training = mean = 45
– Drama Training = mean = 41
– No Training = mean = 42
- adaptive skills (no significant difference among groups pretest)
– Keyboard Training = mean = 57
– Singing Training = mean = 60
– Drama Training = mean = 57
– No Training = mean = 59
- Were the groups similar before intervention began? Yes
- Were the communication problems adequately described? NA. The Ps are assumed to be typically developing.
- Was membership in groups maintained throughout the study?
- Did each of the groups maintain at least 80% of their original members? Yes
- Were data from outliers removed from the study? No
- Were the groups controlled acceptably? Yes
- Was there a no intervention group? Yes
- Was there a foil intervention group? No
- Was there a comparison group? Yes
- Was the time involved in the foil/comparison and the target groups constant? Yes
- Were the outcomes measure appropriate and meaningful? Yes
- The outcomes (dependent variables) were
- OUTCOME #1: Correct responses to English sentences depicting sad – happy affective prosody.
- OUTCOME #2: Correct responses to English sentences depicting fearful – sad affective prosody.
- OUTCOME #3: Correct responses to Tagalog sentences depicting sad – happy affective prosody.
- OUTCOME #4: Correct responses to Tagalog sentences depicting fearful – sad affective prosody.
- OUTCOME #5: Correct responses to tone sequences derived from English sentences depicting sad – happy affective prosody.
- OUTCOME #6: Correct responses to tone sequences derived from English sentences depicting fearful – sad affective prosody.
- OUTCOME #7: Correct responses to tone sequences derived from Tagalog sentences depicting sad – happy affective prosody.
- OUTCOME #8: Correct responses to tone sequences derived from Tagalog sentences depicting fearful – sad affective prosody.
- All the outcome measures were subjective.
- None of the outcome measures were objective.
- Were reliability measures provided?
- Interobserver for analyzers? No
- Intraobserver for analyzers? No
- Treatment fidelity for clinicians? No
- What were the results of the statistical (inferential) testing
- Summary Of Important Results
– What level of significance was required to claim significance? p = 0.05
NOTE: The investigators described their analyses of response patterns rather than listing the findings for each of the outcomes.
– Overall, correct responses were significantly higher for spoken utterances compared to tone sequences.
– Overall, correct responses were significantly higher for English vs Tagalog stimuli.
– Overall, correct responses were significantly higher for happy-sad stimuli than for fearful-angry stimuli.
– Overall, the happy-sad distinctions were at ceiling. Differences among interventions were more likely to emerge in the fearful-angry distinctions.
– The drama and keyboard (but not singing) outperformed the no treatment group.
– The drama and keyboard groups were not significantly different.
– The singing group was not significantly different than the drama group nor was it significantly different from the no treatment group.
– The keyboard group outperformed the singing group.
- What was the statistical test used to determine significance? ANOVA
- Were confidence interval (CI) provided? No
- What is the clinical significance? NA
- Were maintenance data reported? No
- Were generalization data reported? Yes. Since the outcomes/dependent measures were not the focus of the intervention, all the outcomes could be considered generalizations.
- Describe briefly the experimental design of the investigation.
- The Ps were recruited from a previous investigation.
- In the previous investigation, six-year old typically developing children had been assigned to one of four conditions: keyboard, singing, drama, no treatment.
- Ps received 1 year of treatment during 1st grade in the prior investigation.
- Prior to the intervention from the previous investigation, the Ps were tested on a number of measures but not on the outcome measures from the current investigation. The investigators note that there were post test versions of the pretest measures but they are not reported in this investigation.
- After the post tests of the previous investigation, Ps of the previous investigation were invited to participate in the current investigation.
- The investigators analyzed the data using a mixed 4 x 2 x 2 x 2 design: training conditions (keyboard, singing, drama, no treatment; between subjects) x modality (spoken utterances, tone sequences) x language (English, Tagalog) x emotion (sad-happy vs; fearful vs angry)
ASSIGNED OVERALL GRADE FOR QUALITY OF EXTERNAL EVIDENCE: D
SUMMARY OF INTERVENTION
PURPOSE: To identify an intervention that resulted in improved comprehension of affective prosody
POPULATION: Typically developing 1st Graders
MODALITY TARGETED: comprehension
ELEMENTS/FUNCTIONS OF PROSODY TARGETED: affective prosody
ELEMENTS OF PROSODY USED AS INTERVENTION: music (pitch, intonation, rhythm); possibly stress, tempo
DOSAGE: weekly lessons for 1 year, 45 minute, group lessons
ADMINISTRATOR: conservatory affiliates or graduates in drama, singing, or music
- There were 3 possible interventions:
- Traditional piano training using the electronic keyboard.
- Teaching procedures included:
– sight reading
– music notation
– fingering patterns
– playing from memory
– clapping rhythms (p. 56)
- Used the Kodály method which included
– dancing (p. 56)
- Used simple tunes and folk tunes
- Used the scripts of simple plays.
- Teaching procedures included