Mean speech rate across Age, Frequency, and Sex groups

  • The data for this exercise were gathered by Pluymaekers, Ernestus & Baayen (2005), who investigated the duration of the Dutch past participle prefix ge:
    • Infinitive zetten (“to put”)
    • Past participle: gezet
  • Here we will compare the SpeechRate (measured in syllables per second) of four age cohorts (ageCohort):
    • Speakers born 1900-1925
    • Speakers born 1926-1950
    • Speakers born 1951-1975
    • Speakers born 1976-2000
  • We will look at a discretized version of word Frequency, which clusters words in quartile groups in function of their frequency:
    • 1st frequency quartile
    • 2nd frequency quartile
    • 3rd frequency quartile
    • 4th frequency quartile
  • We will also take into account the influence of gender (Sex):
    • Male
    • Female

1. Mean speech rate across ageCohort

1.1 Loading and exploring the data

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

1.2 Computing and plotting group means

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

1.3 Removing outliers

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

1.4 Checking the assumptions of a one-way ANOVA

  • Recall that the assumptions of a one-way ANOVA are the following:
  • The observations are independent
  • The response variable is ratio- or interval-scaled
  • The scores in the groups are normally distributed
  • The variance is homogeneous, i.e., the variances of the different groups should be equal
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

1.5 Performing a one-way ANOVA

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

2. Mean speech rate across Frequency

2.1 Computing and plotting group means

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

2.2 Checking the assumptions of a one-way ANOVA

  • Recall that the assumptions of a one-way ANOVA are the following:
  • The observations are independent
  • The response variable is ratio- or interval-scaled
  • The scores in the groups are normally distributed
  • The variance is homogeneous, i.e., the variances of the different groups should be equal
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

2.3 Performing a one-way ANOVA

eyJsYW5ndWFnZSI6InIiLCJwcmVfZXhlcmNpc2VfY29kZSI6ImxpYnJhcnkocmVhZHIpXG5saWJyYXJ5KGRwbHlyKVxuXG5kYXRhU2V0IDwtIHJlYWRfY3N2KFwiaHR0cDovL3d3dy5qZXJvZW5jbGFlcy5iZS9zdGF0aXN0aWNzX2Zvcl9saW5ndWlzdGljcy9kYXRhc2V0cy9jbGFzczlfUGx1eW1hZWtlcnNfZXRfYWxfMjAwNS5jc3ZcIilcbmRhdGFTZXQgPC0gbXV0YXRlX2lmKGRhdGFTZXQsIGlzLmNoYXJhY3RlciwgYXMuZmFjdG9yKVxuZGF0YVNldDwtZGF0YVNldFthYnMoc2NhbGUoZGF0YVNldCRTcGVlY2hSYXRlKSk8PTIsIF0iLCJzYW1wbGUiOiIjIFRoZSBtb2RpZmllZCBkYXRhLmZyYW1lIGRhdGFTZXQgaXMgYWxyZWFkeSBpbiB5b3VyIHdvcmtzcGFjZVxuXG4jIFNwZWNpZnkgYW4gQU5PVkEgbW9kZWwgb2YgU3BlZWNoUmF0ZSB+IEZyZXF1ZW5jeS4gU3RvcmUgdGhlIG1vZGVsIGluIHRoZSB2YXJpYWJsZSAnbW9kZWwnXG5cbiMgQ29tcHV0ZSBhIHN1bW1hcnkgb2YgdGhlIG1vZGVsIG9iamVjdFxuXG4jIFVzZSBUdWtleUhTRCAoVHVrZXkgSG9uZXN0IFNpZ25pZmljYW50IERpZmZlcmVuY2UpIHRlc3QgdG8gZXN0YWJsaXNoIHdoaWNoIGdyb3VwcyBhcmUgc2lnbmlmaWNhbnRseSBkaWZmZXJlbnQgZnJvbSBlYWNoIG90aGVyXG5cbiMgV2hhdCBkbyB5b3Ugc2VlPyBBcmUgdGhlIGRpZmZlcmVuY2VzIGJldHdlZW4gdGhlIEZyZXF1ZW5jeSBncm91cHMgc2lnbmlmaWNhbnQ/IFdoaWNoIGdyb3VwcyBjb250cmlidXRlIHRvIHRoaXM/IFlvdSB3aWxsIGZpbmQgdGhlIGNvcnJlY3QgYW5zd2VyIG9uIHRoZSBTb2x1dGlvbiB0YWIiLCJzb2x1dGlvbiI6IiMgVGhlIG1vZGlmaWVkIGRhdGEuZnJhbWUgZGF0YVNldCBpcyBhbHJlYWR5IGluIHlvdXIgd29ya3NwYWNlXG5cbiMgU3BlY2lmeSBhbiBBTk9WQSBtb2RlbCAod2l0aCBgYW92YCkgb2YgU3BlZWNoUmF0ZSB+IEZyZXF1ZW5jeS4gU3RvcmUgdGhlIG1vZGVsIGluIHRoZSB2YXJpYWJsZSAnbW9kZWwnXG5tb2RlbDwtYW92KFNwZWVjaFJhdGUgfiBGcmVxdWVuY3ksIGRhdGFTZXQpXG4jIENvbXB1dGUgYSBzdW1tYXJ5IG9mIHRoZSBtb2RlbCBvYmplY3RcbnN1bW1hcnkobW9kZWwpXG4jIFVzZSBUdWtleUhTRCAoVHVrZXkgSG9uZXN0IFNpZ25pZmljYW50IERpZmZlcmVuY2UpIHRlc3QgdG8gZXN0YWJsaXNoIHdoaWNoIGdyb3VwcyBhcmUgc2lnbmlmaWNhbnRseSBkaWZmZXJlbnQgZnJvbSBlYWNoIG90aGVyXG5UdWtleUhTRChtb2RlbClcbiMgV2hhdCBkbyB5b3Ugc2VlPyBBcmUgdGhlIGRpZmZlcmVuY2VzIGJldHdlZW4gdGhlIEZyZXF1ZW5jeSBncm91cHMgc2lnbmlmaWNhbnQ/IFdoaWNoIGdyb3VwcyBjb250cmlidXRlIHRvIHRoaXM/IFlvdSB3aWxsIGZpbmQgdGhlIGNvcnJlY3QgYW5zd2VyIG9uIHRoZSBTb2x1dGlvbiB0YWJcbiMgLSBUaGUgZmFjdG9yIGlzIG5vdCBzaWduaWZpY2FudC4gXG4jIC0gQ2xvc2VyIGluc3BlY3Rpb24gb2YgdGhlIHBhaXJ3aXNlIGdyb3VwIGRpZmZlcmVuY2VzIGNvbmZpcm1zIHRoZSByZXN1bHRzIHdlIG9idGFpbmVkIGJ5IGNvbXBhcmluZyB0aGUgYnktZ3JvdXAgbWVhbnM6IHRoZXJlIGlzIG5vdCBtdWNoIGRpZmZlcmVuY2UgYmV0d2VlbiB0aGUgRnJlcXVlbmN5IGdyb3VwcyAgIiwic2N0IjoidGVzdF9vdXRwdXRfY29udGFpbnMoXCJzdW1tYXJ5KG1vZGVsKVwiLCAgaW5jb3JyZWN0X21zZyA9IFwiTWFrZSBzdXJlIHRvIGNvbXB1dGUgYSBzdW1tYXJ5IG9mIHRoZSBhb3YgbW9kZWwhXCIpXG50ZXN0X291dHB1dF9jb250YWlucyhcIlR1a2V5SFNEKG1vZGVsKVwiLCBcIkRvbid0IGZvcmdldCB0byBwZXJmb3JtIHRoZSBUdWtleUhTRCFcIilcblxuc3VjY2Vzc19tc2coXCJJIHRoaW5rIHlvdSBkaWQgZ3JlYXQhXCIpIn0=

3. Mean speech rate across Sex

3.1 Computing and plotting group means

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

3.2 Checking the assumptions of a one-way ANOVA

  • Recall that the assumptions of a one-way ANOVA are the following:
  • The observations are independent
  • The response variable is ratio- or interval-scaled
  • The scores in the groups are normally distributed
  • The variance is homogeneous, i.e., the variances of the different groups should be equal
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

3.3 Performing a one-way ANOVA

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

4. Two-way anova: the influence of ageCohort and Frequency on SpeechRate

4.1 Computing and plotting group means

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

4.2 Checking the assumptions of a two-way ANOVA

  • Recall that the assumptions of a two-way ANOVA are the following:
  • The observations are independent
  • The response variable is ratio- or interval-scaled
  • The scores in the groups are normally distributed
  • The variance is homogeneous, i.e., the variances of the different groups should be equal
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

4.3 Performing a two-way ANOVA

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

5. Two-way anova: the influence of ageCohort and Sex on SpeechRate

5.1 Computing and plotting group means

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

5.2 Checking the assumptions of a two-way ANOVA

  • Recall that the assumptions of a two-way ANOVA are the following:
  • The observations are independent
  • The response variable is ratio- or interval-scaled
  • The scores in the groups are normally distributed
  • The variance is homogeneous, i.e., the variances of the different groups should be equal
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

5.3 Performing a two-way ANOVA

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

References

Pluymaekers, M., Ernestus, M. and Baayen, R. H. (2005) Frequency and acoustic length: the case of derivational affixes in Dutch, Journal of the Acoustical Society of America, 118, 2561-2569.

© 2018 Jeroen Claes