Vous êtes sur la page 1sur 3

What are the Iowa Assessments?

A tests validity refers to the extent to which it actually measures what it


purports to measure. According to the Iowa Assessments Research and
Development Guide, The IOWA Assessments are developed by the faculty
and staff at Iowa Testing Programs (ITP) at The University of Iowa. They are
large-scale achievement tests that assess students foundational and higher
order thinking skills in a number of academic areas. They have served as an
invaluable part of the research program in educational measurement at The
University of Iowa for over 80 years. Results of these assessments can help
teachers monitor the progress of individual students or an entire class. They
can also inform instructional decisions in order to increase students learning
(p. 3).
How were the Iowa Assessments piloted and normed?
Norms compare one students scores with those obtained by other students
in the same age or grade range. Comparisons of this nature allow teachers
and administrators to assess the performance of their students relative to
that of a nationally representative sample. (p. 11)

The creators of the Iowa Assessments administered the tests to large groups
of students nationwide. They collected data from a number of different
populations including public schools, and private schools- both secular and
non-secular. They did this to ensure that the norming sample mirrored the
national population as closely as possible and insured proportional
representation of important groups of students. In selecting their National
Comparison Sample, they sought to obtain a national probability sample
representative of students nationwide, a nationwide sample of schools for
school-building norms, and data for Catholic/private (non-Catholic) norms
and other special norms. (p. 11-12)

Private schools were considered in terms of whether they were Catholic or


non-Catholic. In order to obtain a properly representative sample of public
school students, developers of the Iowa Assessments classified schools
according to three main stratifying variables geographic region, district
enrollment, and Title I status (or socioeconomic status). They grouped each
combination of the three variables (e.g., Midwest, high enrollment, receiving
Title I funds) and selected school buildings randomly from that group. In

collecting data for the Catholic School sample, researchers used the National
Catholic Education Association (NCEA)/Ganleys (2010) Catholic Schools in
America as their primary source for selecting weighting schools. Schools
were separated into the same geographic regions used in obtaining the
public school sample, and then stratified into five categories based on
diocesan enrollment. From there, a two-stage random sampling process was
used to select the sample. The sample of non-Catholic schools was obtained
from the Quality Education Data (QED) file. Schools were sampled from each
geographic region until the targeted numbers of students were reached. (p.
12-13)

During the 2010 research study, each participant was administered the
appropriate level (5/6 through 17/18) of the Iowa Assessments. Some
students were also administered additional tests such as the Cognitive
Abilities TestTM (CogAT) or the Iowa Test of Basic Skills (ITBS ). Once all
testing materials were received by Riverside Scoring ServiceTM, the number
and percentages of students in each sample and stratification category were
determined. Percentages were adjusted by weighting to compensate for
categories that were not represented and to correct for schools that tested a
different number of students than was required. This allowed for weighted
distributions to be calculated so that the three comparison samples could be
considered to closely approximate the total student population. (p. 13)

Reliability and Validity of the Iowa Assessments


The reliability of an assessment refers to the likelihood that multiple trials by
the same individual would yield the same results. In order to confirm the
reliability of the assessment during development, the developers of the Iowa
Assessments employed two different methods.
First they utilized internal consistency estimates using Kuder-Richardson
Formula 20 (K-R 20). This derives reliability coefficients based on data from
the whole national comparison sample using scores reported in both the fall
and the spring administrations.
The second method that was used estimated reliability based on two testing
occasions. Although this version of the Iowa Assessments was published in
2013, developers measured alternate-forms reliability in 2010 using Form E
of the Iowa and Form A of the ITBS/ITED because these forms were equated

in the fall of 2010. They were also able to estimate test-retest reliability
using a comparability study done in 2011-2012 between paper-based and
computer-based administrations of Form E of the Iowa. (p. 70)

A tests validity refers to the extent to which it actually measures what it


purports to measure. According to the Iowa Assessments Research and
Development Guide, the most valid assessment of achievement for a
particular school is one that most closely defines the schools education
standards and goals for teaching and learning, (p. 4). The assessment
framework for the Iowa Assessments relies on a variety of sources to confirm
and support its content validity, including:

State, professional, and international standards


Curriculum surveys
National Assessment of Educational Progress (NAEP) frameworks and
test specifications
Scholarly research
Feedback from educators, parents, and students
Assessment data

Reference:
Iowa Assessments Research and Development Guide:
http://help.riversidepublishing.com:8080/robohelp/robo/server/DM_DigRes_D
ev/projects/DM_Digital_Resources/Baggage_Files/Iowa/Iowa_E_F_RnD_Guide_v
1_SEC.pdf

Vous aimerez peut-être aussi