 |
|
 |
Issue 19: December 2008 Helping Hands/Warm Hands One of the great pleasures in life is in the act of giving to another. The Improve Group is very pleased to have the opportunity to help spread some warmth and support to worthy programs in our community. In November we chose three special organizations to receive grants from our corporate giving program. The three recipients are as follows: The innovative new program from Project for Pride in Living (PPL), Cre8_Peace, will engage young people in using technology to reduce community violence. Heart of the Beast Puppet and Mask Theater will be providing artistic support. The young people participating have strongly expressed an interest in using technology to learn, be creative, and make a difference in their community. The $1,000 grant from the Improve Group will be used to purchase digital cameras and art supplies essential to the program. (photo) PPL working with youth in the CRE8_Peace program St. Anne’s Place, an emergency shelter located in North Minneapolis, helps women and their families gain stability through support in a proactive, nurturing environment. The $1,000 grant from the Improve Group will help St. Anne’s place purchase sturdy educational toys that its youngest residents can play with to learn, as well as relax, have fun and maintain a sense of normalcy during a stressful period in their family’s life. The Improve Group’s gift of $1,000 to the Page Education Foundation will support one student scholarship for the current academic year. Page Education Foundation fosters educational success, personal responsibility and the full potential of young people of color. Each of its college-age scholarship recipients mentors school-age students to serve as role models and inspire high educational aspirations and achievement.
(photo) Leah delivering a check to staff and scholarship recipients from the Page Education Foundation We made another special donation this year in honor of the clients and colleagues who allowed us to work closely with them – and brought us many laughs! - throughout the year. Twenty pairs each of “Mittens with a Mission” were donated to three charities nominated by the IG staff. Mittens with a Mission are beautiful wool and cotton creations put together at the Winona Occupational Rehabilitation Center. The charities receiving mittens were:
Emma Norton Services was suggested by Danielle, a research analyst with the Improve Group. The mission of Emma Norton is to foster stability and personal growth for women and families who are homeless- or at risk of becoming homeless- by providing affordable, secure housing with additional support, including school, work or a self-improvement program of their choosing. Residents can live in the Emma Norton Residence for up to two years while they work towards their goal of independence and self-sufficiency. We thought that this would be a great place to help keep hands warm during the long winter months! St. Joseph’s Home for Children helps children and their families in crisis. Children can come to St. Joseph’s for the day or to stay for treatment when experiencing severe emotional or behavioral issues. St. Joseph’s also provides emergency shelter for children who are at risk of abuse or neglect.
|
|
Bias in Pre- and Post-Test Surveys At the Improve Group, we work with surveys and survey data nearly every day. Surveys are a valuable tool in evaluation to understand how people think, feel or behave. As with any other evaluation tool, surveys have limitations. This article focuses on two types of bias that can occur when using surveys, and that you should be aware of when analyzing and sharing data. Traditional pre/post-test surveys are administered once before an intervention (or program) starts and again when it ends. The traditional pre /post-test survey method is subject to response-shift bias, a bias that occurs when survey respondents realize “what they did not know” only after they are exposed to an intervention. In other words, survey respondents may overestimate their knowledge or ability at pre-test. Then, after participating in a program, respondents rate themselves more accurately at post-test. When reviewing data, the change from the beginning to the end of the program is under-reported because the respondent overestimated their knowledge or ability at pre-test. The retrospective pre/post-test design is used to address the response-shift bias in traditional pre/post-test surveys. The retrospective pre/post test is administered only once, after participation in the program. Respondents are asked to rate their opinion or experience at the time of the survey as well as at the beginning of the program. The retrospective pre-test format allows respondents to rate their change after having experienced the intervention. We have found that the retrospective pre-test/post-test design works well to measure the changes respondents experience from a program. However, in our studies that employ comparison groups, we noticed another related bias that we have coined “exposure bias.” Exposure bias occurs when one group of respondents is experiencing response-shift bias and the other group is not. In other words, while the treatment group learns “what they did not know” through exposure to the program, the comparison group never learns “what they do not know.” In our work with the Performing Arts Workshop, we examine how their program (which provides teaching artist residencies to special education classrooms) impacts students and teachers. We use a quasi-experimental design; we survey treatment classrooms that receive the program and comparison classrooms that do not receive the program. We recently found that teachers in comparison classrooms consistently rated themselves more favorably at pre-test than teachers in treatment classrooms on items such as teachers’ level of comfort using arts in the classroom (See Figure 1 below). This indicates that teachers in treatment classrooms may have assessed their comfort level in teaching the arts by thinking of the expert teaching artist as their standard for comparison. In contrast, teachers in comparison classrooms did not work with a teaching artist in the classroom and so did not hold themselves to such a high standard. We have found that while exposure bias is inevitable in a quasi-experimental research design, we can take steps to mitigate its impact on our findings. To address the exposure bias, we compare change over time from pre-to post-test, instead of just looking at the post-test score. This helps us examine program impact while accounting for exposure bias. Observations, interviews and focus groups also help us to understand our findings. Figure 1: Teachers’ level of comfort for using dance and creative movement in their classroom 
|
|
|
Holiday Creativity at the Improve Group
|
|
mali Community in Minnesota
