SELF-SELECTION AND THE FLSR - PowerPoint PPT Presentation

About This Presentation
Title:

SELF-SELECTION AND THE FLSR

Description:

You re messing up our results! SELF-SELECTION AND THE FLSR ANN LEWIS Background The FLSR program at BYU Contact with native speakers Minimum participation through ... – PowerPoint PPT presentation

Number of Views:26
Avg rating:3.0/5.0
Slides: 6
Provided by: OpenAcces5
Learn more at: https://www.ttt.org
Category:
Tags: and | flsr | selection | self | the | sentences

less

Transcript and Presenter's Notes

Title: SELF-SELECTION AND THE FLSR


1
Youre messing up our results!
  • SELF-SELECTION AND THE FLSR
  • ANN LEWIS

2
Background
  • The FLSR program at BYU
  • Contact with native speakers
  • Minimum participation through dinners and church
  • Alleged full immersion in language use in-home
  • Middle ground between traditional classroom and
    study abroad
  • Less expensive to student
  • Evaluation
  • Examine pronunciation and fluency over time
  • Compare FLSR and non-FLSR students to
    native speakers
  • Relation to linguistic trends
  • Study back vowels
  • Study bilabial consonants

3
Two Sides a Subissue and Stumbling Block
  • Self-selection Bias a possible confound in
    research in which a subject has the choice of
    participation.
  • Side A Students who volunteer for challenging
    academic programs are high-initiative and likely
    have already learned more than their classmates.
  • Side B Language aptitude is inherent. One can
    study for years and never qualitatively improve.
    Also, social factors are involved that are
    unrelated to language skill and will likely
    weaken any intellectual effect.

4
Methodology
  • Non-equivalent matched groups control for age,
    gender, years of study, mission, and current
    class level
  • FLSR Residents
  • Non-FLSR Spanish students
  • Affect-neutral sentences containing back vowels
    and bilabial consonants in stressed and
    unstressed positions (/u/, /o/, /a/, /p/, and
    /B/)
  • Record sentences in as conversational of a manner
    as possible
  • Measure F1, F2, F3, duration, VOT, frication, and
    frequency 
  • Normalize formants to facilitate cross-gender
    comparison
  • Run ANOVAs for measurements comparing gender and
    group as independent variables
  • (Full project)
  • 10 weeks later
  • Similar affect-neutral sentences to adjust for
    practice effect (words within 150 on Davies
    Frequency List)
  • All retested (natives included to adjust for
    drift) 
  • Measurements retaken and normalized
  • Run ANOVAs on difference between pre- and
    post-test measures
  • Native Spanish speakers evaluate the anonymous
    recordings for accentedness, comprehensibility,
    and fluency on 7-pt Likert scales.
  • Run ANOVAs on group means

5
Possible Problems
  • Within-groups variance should be negated by
    within-subjects measures
  • 34 sounds per person
  • 4 measurements per sound
  • Test as a measure of reading
  • Stuttering, formal pronunciation
  • Unpredictable phonotactics
  • Swallowed sounds
  • Nasalization of vowels (único)
  • Self-selection over time
  • Small view of a large issue
  • But in the future
  • Statistical evaluation of fluency factors
  • Intonation
  • Intensity
  • Stress
  • Other phonemes
Write a Comment
User Comments (0)
About PowerShow.com