SPATIAL SKILLS TRAINING TO IMPROVE STUDENT SUCCESS IN ENGINEERING - - PowerPoint PPT Presentation

SPATIAL SKILLS TRAINING TO IMPROVE STUDENT SUCCESS IN ENGINEERING

Sheryl A. Sorby Visiting Professor The Ohio State University Professor Emerita Michigan T echnological University

ENGINEERING IS ONE OF THE MOST SPATIALLY DEMANDING FIELDS

Source: Johnson O’Connor Research Foundation

SPATIAL SKILLS AND ENGINEERING GRAPHICS

A 1985 study at Michigan Tech determined

that a student’s score on the PSVT:R was the most significant predictor of success of eleven variables tested in engineering graphics

EXAMPLE: EASE OF LEARNING CAD SOFTWARE

Study conducted in Fall 2006 with first-year engineering

students

Students were pre-tested with a test designed to assess

3-D spatial skills

 Divided into three groups: High Visualizers, Average Visualizers, and Low

Visualizers

CAD software instruction module followed At end of instructional module, students were given a

survey regarding their perceptions of learning the software

STUDY RESULTS (5-POINT SCALE—HIGHER NUMBERS=MORE DIFFICULTY) (N=329)

Low Avg High

Confidence in starting assignment

2.75 2.12 1.91

Ease in planning modeling approach

2.64 2.08 1.91

Time spent modeling part

2.29 1.81 1.84

Number of times starting over

2.04 1.69 1.65

Ease of working with software

3.02 2.64 2.73

Amount of Assistance required

2.33 2.17 1.77

Ease in learning compared to teammates

2.87 2.49 2.35

BACKGROUND FACTORS IN THE DEVELOPMENT OF SPATIAL SKILLS FOR ENGINEERING STUDENTS

Play with construction toys Shop, drafting, and mechanics classes 3-D computer games Certain sports Mathematics skills Sketching Most factors in developing skills have a certain

degree of gender bias favoring males

GENDER DIFFERENCES ARE ROBUST AND CONSISTENT

RECENT DATA ANALYSIS BY RACE/ETHNICITY

Male Female n Average n Average African American 93 19.957 p<0.0001 38 13.816 p<0.0001 American Indian 63 22.619 p<0.0002 18 20.222 NS Asian American 66 23.530 p=.05 30 19.333 NS Hispanic 97 24.227 NS 20 18.700 NS Multiracial 88 24.239 NS 27 20.481 NS White 7562 24.488 1441 20.712

RECENT DATA ANALYSIS BY COUNTRY OF ORIGIN

n Average Standard Deviation

• Sig. of

Differences Middle East 16 15.563 5.921 p<0.0001 Africa 28 16.714 5.199 p<0.0001 India 47 14.872 5.859 p<0.0001 China 82 22.268 4.456 p=0.0008 Far East Asia 27 23.630 5.534 NS South America 16 22.438 4.273 NS Canada 11 24.545 3.804 NS Other 15 22.133 5.167 NS

ENGINEERING DEMOGRAPHICS

Engineering is one of the least diverse STEM fields ~ 18% women ~5% African American ~5% Hispanics The lack of diversity in engineering could be linked

to poorly developed spatial skills for these students

SPATIAL SKILLS REMEDIATION

 A 1-credit course in developing spatial skills has been offered at Michigan T

ech since 1993

 Course has been adopted at several other engineering schools across the

country in recent years

 Purdue  Ohio State  Virginia T

ech

 UT Austin  University of Illinois-Chicago  Others  Goal is to improve spatial skills so that students are more successful in

engineering

SPATIAL SKILLS COURSE FORMAT

One 1.5-hour lab session per week for one credit Short mini-lecture (~10-15 minutes) at beginning of

session

Students work through software module in teams of two Students complete workbook pages for remainder of

time

COURSE MODULES

Isometric Sketching Orthographic Projection

with Normal Surfaces

Orthographic Projection

with Inclined and Curved Surfaces

Flat Pattern Developments Rotation of Objects about a

Single Axis

Rotation of Objects about

Two or more Axes

Object Reflection and

Symmetry

Cross-Sections Surfaces and Solids of

Revolution

Combining Objects

MULTIMEDIA SOFTWARE AND WORKBOOK

Each software module has a background section

and an exercise section

Exercises are primarily matching or fill in the

blank

Workbook modules include background

information as well as multiple exercises

Additional Sketching exercises available in the

workbook

MULTIMEDIA SOFTWARE

MODULE 1: ISOMETRIC SKETCHING

IMAGINE A DIAGONAL FOR THE CUBE…

DIFFERENCE BETWEEN 3-D VIEW AND ISOMETRIC VIEW

PSVT:R TEST RESULTS

20 40 60 80 100 1993 1994 1995 1996 1997 1998

Pre-Test Post-Test

Gains statistically significant

ASSESSMENT OF GAINS ON SPATIAL SKILLS TEST-PSVT:R

n Pre- Test Post- Test Gain Significance of Gain Original Course 186 50.5 76.9 26.4 p<0.0001 Modified Course 157 48.3 73.7 25.4 p<0.0001

LOGISTIC REGRESSION RESULTS FROM RESEARCH WITH NON-ENGINEERING MAJORS

 Both groups that used the workbook were significantly better than the

CG

 Software only group was not better than CG

FOR EXAMPLE: AVERAGE GPAS IN COURSES 1996-98 AND 2000-02

Course Failed PSVT:R, enrolled Passed PSVT:R with score 60-70% Failed PSVT:R, did not enroll Pre-Calculus 2.74 2.41 p=0.002 2.19 p<0.0001 Calculus I 2.59 2.48 N.S. 2.25 p=0.004 Chemistry I 2.64 2.47 p=0.02 2.31 p=0.004 Computer Science I 3.16 2.88 p=0.02 2.53 p<0.0001 Overall 3.01 2.84 p=0.0005 2.63 p<0.0001

FALL 2009 DATA—NO SELF SELECTION

Course Passed PSVT:R score

• f 60-70%

Failed PSVT:R and took spatial skills course Significance of Difference

Pre-Calculus 2.06 (s=1.093, n=62) 2.23 (s=1.61, n=61) NS Calculus I 2.27 (s=1.384, n=120) 2.63 (s=1.323, n=106) p=0.024 Chemistry I 2.35 (s=1.061, n=149) 2.51 (s=0.946, n=129) p=0.096 Computer Science I 2.25 (s=1.356, n=20) 2.63 (s=1.008, n=16) NS Overall GPA 2.64 (s=0.907, n=199) 2.83 (s=0.726, n=187) p=0.012

GRADUATION RATES (STUDENTS MATRICULATING 1996-1998)

10 20 30 40 50 60 70 80 90 At University Within Engineering Passed PSVT:R 70%

• r

higher Failed-Took course Passed PSVT:R 60-70% Failed-Did not take course

RETENTION RATES AT UNIVERSITY BY GENDER

Women Men

Failed PSVT:R- Took course Failed PSVT:R- Did not take course Passed PSVT:R score of 60% or higher Failed PSVT:R- Took course Failed PSVT:R- Did not take course Passed PSVT:R score of 60% or higher

Retention Rate 2000-2002 87.4% 71.1% 83.2% 76.8% 70.0% 73.4% Retention Rate 1993- 1998 88.9% 68.3% 87.2% 75.3% 69.0% 72.4%

QUESTIONS?