A Meta-Analysis on the Effects of Bilingual Education Programs for Dual Language Learners' Academic Achievement.

Megan Groth Author

04/01/2021 Added

30 Plays

Description

This presentation demonstrates the process of running a meta-analysis on the effects of bilingual education programs for dual language learners academic achievement

Searchable Transcript

Search:

[00:00:02.520]Hi, my name is Megan Groth,
[00:00:04.160]and the research project I worked on is
[00:00:06.330]A Meta-Analysis on the Effects
[00:00:08.090]of Bilingual Education Programs
[00:00:10.520]for Dual Language Learner's Academic Achievement.
[00:00:13.440]My Research Advisor for this project is Dr. Marc Goodrich.
[00:00:18.300]Schools have dual language programs
[00:00:19.920]to help kids receive access to educational content
[00:00:22.900]and educational instruction in the child's home language,
[00:00:26.290]which is not English,
[00:00:27.440]by providing instruction in their home language
[00:00:29.570]while they are still developing English proficiency.
[00:00:33.060]Other potential outcomes are improving skills in the home
[00:00:36.410]and maintaining kids' connection
[00:00:37.800]to their heritage and culture.
[00:00:39.810]It can be a very culturally sensitive approach to education.
[00:00:43.420]However, there's not very much empirical evidence
[00:00:46.050]examining how effective dual language programs are.
[00:00:50.760]The purpose of this meta-analysis is to determine
[00:00:53.500]if bilingual education programs result
[00:00:55.960]in improved outcomes for dual language learners.
[00:00:59.070]Prior findings are mixed,
[00:01:00.320]and prior meta-analysis include poor quality studies,
[00:01:03.710]so we are not able to draw conclusions.
[00:01:06.520]The objective was to run a meta-analysis
[00:01:08.660]to determine what the average effect
[00:01:10.370]of these different programs had
[00:01:12.150]on dual language learner's academic outcomes.
[00:01:15.070]We hypothesize that children in dual language programs
[00:01:17.890]would have lower English proficiency
[00:01:20.110]than children in monolingual programs
[00:01:22.380]because they are exposed to English for less time.
[00:01:25.350]However, because evidence suggests
[00:01:27.650]that children can transfer academic knowledge
[00:01:29.870]and skills across languages,
[00:01:31.830]we expected that children in dual language programs
[00:01:34.380]would do as well or better
[00:01:36.000]than children in monolingual programs
[00:01:38.330]on academic outcomes such as reading and math.
[00:01:41.640]We also examined whether the effectiveness
[00:01:43.580]of dual language programs change
[00:01:45.510]depending on the specific type of program a school used.
[00:01:49.500]For example, some schools use transitional programs,
[00:01:52.530]in which children start with instruction
[00:01:54.560]in their home language
[00:01:55.920]but transition to English by the third grade
[00:01:58.810]and other schools use 50-50 immersion programs,
[00:02:01.470]in which half of the instruction is in the home language
[00:02:04.410]and half of the instruction is in English
[00:02:06.250]throughout all of elementary school.
[00:02:10.150]The methods process was broken down into three steps,
[00:02:12.930]literature search, determining inclusion and exclusion,
[00:02:15.970]and coding.
[00:02:17.320]The first step was the literature search.
[00:02:19.630]The author used PsychInfo,
[00:02:21.500]ProQuest Dissertations Thesis A and I,
[00:02:24.750]and ERIC to search for published articles, book chapters,
[00:02:27.940]dissertations, and thesis.
[00:02:30.470]PsychInfo searched for psychology related articles,
[00:02:33.700]and ERIC searched for education related articles.
[00:02:37.470]The author then used key terms,
[00:02:39.070]which are listed on the right of the slide,
[00:02:40.780]to narrow down the search.
[00:02:43.910]The second step was to determine inclusion and exclusion.
[00:02:47.390]The following stipulations were made for inclusion.
[00:02:50.580]All studies must have been available on the online database
[00:02:53.770]no later than January 15th, 2021, and written in English.
[00:02:58.980]Studies must include dual language learners,
[00:03:02.640]a bilingual education program,
[00:03:04.820]and academic outcome measures.
[00:03:07.340]Studies must include
[00:03:08.310]typical developing dual language learners, so, for example,
[00:03:11.790]a study of dual language learners with autism
[00:03:14.140]were not included.
[00:03:16.070]Studies that were conducted as experimental,
[00:03:18.530]quasi-experimental,
[00:03:19.790]or ex-post facto group design studies
[00:03:22.620]that measure quantitative outcome variables were included.
[00:03:26.600]Studies needed to report effect size
[00:03:28.500]or the appropriate data for calculating effect size.
[00:03:32.160]The following stipulations were made for exclusion.
[00:03:35.250]Studies that only focused on dual language learners
[00:03:37.730]with learning disabilities
[00:03:39.040]or a physical diagnosis were not included.
[00:03:42.130]Studies that were conducted non-experimental, case studies,
[00:03:45.230]and qualitative studies were also not included.
[00:03:49.500]The third step was coding and reliability.
[00:03:52.770]We coded information about the participants,
[00:03:55.360]the type of dual language program,
[00:03:57.380]languages used in the programs, and outcome measures.
[00:04:00.810]As you can see from the chart on the slide,
[00:04:02.760]this is how the data was recorded.
[00:04:05.240]Other important information coded was
[00:04:07.170]the age and grade of the participants,
[00:04:09.400]the mean and standard deviation of the outcome variables,
[00:04:12.610]and the income and education of the mother and father.
[00:04:16.450]The inter-rater reliability is 98.75,
[00:04:20.170]meaning we had a high agreement
[00:04:21.540]on coding all of the articles.
[00:04:25.880]Now on to the data and results of the study.
[00:04:28.590]Looking at the chart for average effect size,
[00:04:30.890]it shows that,
[00:04:31.723]when averaging across subject areas and program type,
[00:04:34.840]there is no difference in outcomes for children
[00:04:36.950]in dual language and monolingual programs.
[00:04:39.910]In other words,
[00:04:40.743]dual language programs are just as effective
[00:04:42.960]at promoting academic achievement
[00:04:44.600]for dual language learners as are programs
[00:04:47.080]that focus exclusively on development
[00:04:49.570]of language and knowledge in English.
[00:04:53.300]Now here we see a significant moderation by program type,
[00:04:56.710]indicating that different types of dual language programs
[00:04:59.390]result in different outcomes.
[00:05:01.600]Across all subject areas,
[00:05:03.040]kids in dual language programs are scoring
[00:05:05.080]about a quarter of a standard deviation higher
[00:05:07.730]when compared to kids in monolingual programs.
[00:05:10.950]Kids in transitional bilingual programs performed,
[00:05:13.470]on average, worse than their peers in monolingual programs,
[00:05:17.170]and kids in Spanish immersion programs
[00:05:19.090]performed better than their peers in a monolingual program,
[00:05:22.210]which was the most shocking of these results,
[00:05:24.100]and we'll talk about that in our discussion.
[00:05:28.670]Now we look at the different academic subjects.
[00:05:31.020]So if we look at the estimate column for mathematics,
[00:05:34.100]we can see that kids in the bilingual program
[00:05:36.250]performed, on average, better on mathematics
[00:05:39.270]than those in a monolingual program on mathematics.
[00:05:42.430]There were no other differences across academic outcomes.
[00:05:47.050]Then we move to grade,
[00:05:48.190]and if we look at the p-value column,
[00:05:50.110]we see that neither value is significant,
[00:05:52.510]which means there's no difference between the two.
[00:05:56.870]Lastly, we look at publication status,
[00:05:59.520]even though the not published estimate is negative
[00:06:02.620]since the p-value is not significant,
[00:06:04.780]there's no difference found between outcomes
[00:06:07.000]when looking at publication status had an effect.
[00:06:10.140]This suggests there might not be very much publication bias
[00:06:13.130]in the literature
[00:06:14.360]on effectiveness of dual language programs.
[00:06:18.570]In conclusion,
[00:06:19.403]dual language learners did as well or better
[00:06:21.830]as their monolingual peers in every subject, but language.
[00:06:25.570]This makes sense because,
[00:06:26.720]if kids are in Spanish school all day,
[00:06:29.260]English vocabulary will not be as strong,
[00:06:31.950]but given opportunity to acquire academic skills like,
[00:06:34.990]reading, writing, and math in their home language
[00:06:37.790]is still benefiting them
[00:06:39.670]when they go to learn those skills in English.
[00:06:43.170]Earlier I discussed how Spanish immersion
[00:06:45.150]had the biggest impact on Spanish outcomes.
[00:06:47.830]We saw that many of the research articles
[00:06:49.640]looked at Spanish outcomes instead of English outcomes.
[00:06:54.010]This kind of research has important implications
[00:06:56.430]for local, state, and federal education policies
[00:06:59.370]with regard to funding
[00:07:00.420]for school language instruction programs.
[00:07:03.110]Our findings suggest these programs
[00:07:04.900]are not detrimental to children's academic outcomes
[00:07:07.740]and are actually supportive of their use.
[00:07:10.560]I would like to thank
[00:07:11.393]the University of Nebraska Lincoln UCARE Program
[00:07:13.560]for the funding, my advisor, Dr. Marc Goodrich,
[00:07:17.280]and graduate student Lauren Theyer
[00:07:18.850]for allowing me to participate in this project.
[00:07:21.390]Thank you.

The screen size you are trying to search captions on is too small!

You can always jump over to MediaHub and check it out there.

Comments

0 Comments

A Meta-Analysis on the Effects of Bilingual Education Programs for Dual Language Learners' Academic Achievement.

Description

Searchable Transcript

Comments icon comment

Related Channels

Comments