Growth and Achievement Trends of Advanced Placement
2 General Document comments
4 Sentence and Paragraph comments
0 Image and Video comments
1 American Secondary Education 43(2) Spring 2015
2
3 Growth and Achievement Trends of
4 Advanced Placement (AP) Exams in
5 American High Schools
6 Authors
7
8 Eugene Judson, Ed.D., is an Associate Professor of Science Education for the Mary Lou Fulton Teachers College at Arizona State University in Mesa, Arizona.
9 Angela Hobson is a Ph.D. student and Program Evaluator for the Mary Lou Fulton Teachers College at Arizona State University in Mesa, Arizona.
10 Abstract
11
12 This exploratory study examined and compared overall trends in growth and student achievement of the Advanced Placement (AP) program. Using data from the past two decades, analyses indicated there has been steady and extensive growth of AP participation, particularly among underclassmen and some minority groups. However, overall achievement, as measured by pass rates, has declined somewhat over the last 20 years with the proportion of students earning the lowest AP score of a 1 increasing twofold. Possible under-pinning causes discussed include limited course choices for high ability stu-dents, in-class incentive recruitment practices, and changing ethos about AP.
13 Background
14 Advanced Placement (AP) courses in American high schools have been crop-ping up at a tremendous rate in recent years. Correspondingly, the number of students signing up to take AP exams has exploded. From 1992 to 2012, the number of schools offering AP courses nearly doubled and the number of AP exams administered swelled more than 500 percent. The impetus promoting AP courses likely emerges from multiple sources. First, there is a common-sense argument that AP courses allow high school students to get a leg up on college by earning university credit. AP exams are scored on a 1 to 5 scale and typically a score of 3 or higher is considered a passing score that will be accepted for course credit by many colleges across the country. AP courses
16 American Secondary Education 43(2) Spring 2015
17
18 Growth & Achievement TrendsJudson & Hobson
19 are also often viewed as hallmarks of a rigorous curriculum and AP enroll-ment is considered very appropriate for high achieving students. Addition-ally, college admission boards will often view AP success as a predictor of later success in a university setting.
20 Yet there is concern that the United States has gotten ahead of itself (Koebler, 2012). Since the 1990s, the growth of charter schools and the standards-based movement have led to a push for high schools to compete for student enrollment. For high schools, this competition has often led to publicizing a rigorous curriculum chock-full of AP options. AP courses pro-vide the enticements of challenging coursework and potential college credit. However, the passion to promote rigor through AP has possibly created zeal-ousness among schools. In turn, this enthusiasm may be leading to over re-cruitment of students into AP courses.
21 This report describes an examination of AP exam-taking data as well as AP achievement data. The intent was to evaluate the growth of AP over the last two decades and assess how achievement has changed over a twenty year period. The data were examined as a whole, but were also disaggre-gated and scrutinized based on grade levels and ethnic groups. Data were grouped by grade levels because AP growth among the underclassmen has far exceeded that of upperclassmen, so it was a natural inquiry to determine achievement patterns of these students. The rationale for examining growth and achievement patterns based on ethnicity was to scrutinize the alignment with the national agenda to provide more inclusive rigorous curriculum (Bur-ton et al., 2002).
22 Relevant Literature
23 The tremendous growth of AP courses and AP exam-taking has been well documented by both the College Board and researchers (e.g., Farkas & Duf-fett, 2009; Sadler, Sonnert, Tai, & Klopfenstein, 2010). Yet, Tai (2008) point-ed out that, despite continual and steady growth, the AP program has been largely above reproach. Though many consider the expansion of AP as a positive sign of an intensifying focus on rigorous curriculum, others have probed its merits and questioned whether AP has become a proxy for rigor (Sadler, 2010). Iatarola, Conger, and Long (2011) summarized the likeliest causes behind AP growth as being (a) the lure of college course credit that may lower both college costs and time needed to complete a college degree,
24 (b) potential scholarships for high scoring students, and (c) school account-ability calculations tied to AP exams in some states. The establishment of AP courses may also be attributed at least partly to administrators trying to keep high performing students from transferring schools, and attempting to please high performing teachers who prefer high ability students (Iatarola et al., 2010). However, some of these assumptions have been called into ques-tion. For example, in a study of over 28,000 students, Klopfenstein (2010)
26 American Secondary Education 43(2) Spring 2015
27
28 Judson & HobsonGrowth & Achievement Trends
29 found that although AP success does increase the likelihood of completing college in three years for a very small proportion of students, it does not raise the likelihood of graduating from college in four years. Success in dual enrollment high school courses (i.e., yielding secondary and college credit), however, were found to actually increase the likelihood to graduate from college in three, four, and five years.
30 Despite ambiguous results regarding value, AP programs have contin-ued to expand. Lichten (2010) illustrated what he labeled the AP surge with case studies of AP classes and entire high schools where no students earned a passing score of a 3, and yet AP programs continued to grow in these places. Though we clearly know that AP testing has increased tremendously, it is important to also assess if achievement has kept pace. Parents and stu-dents struggle with decisions about enrolling in AP courses, while educators and administrators deliberate whether or not to launch new AP courses. In the context of tremendous AP growth, there has been recent concern that excessive AP course-taking pushes students too hard and that schools may be throttling up too quickly (Koebler, 2012). Additionally, there have been warn-ings that AP course load does not equate to rigor (Challenge Success, 2013).
31 If administrators consider limiting their school’s AP offerings, they may feel pressure that their students will not be as competitive when they apply to top tier universities that have admissions policies favoring AP completion. The banner of being a school that has a student body with a high proportion of AP test-takers does not go unnoticed. Quite notably, the national rank-ing system devised by U.S. News and World Reports to determine lists of “Best High Schools” strongly integrates AP and International Baccalaureate (IB) participation and achievement. In addition to a criterion that identifies high schools that perform better than expected on state accountability as-sessments for all students and for least advantaged students, the U.S. News formula bases selection on calculating schools’ college-readiness perfor-mance. This college-readiness performance criterion uses AP or IB test data, dependent on which program is largest at a school. It is safe to assume that AP is most often the examined program; for example, in 2009 there were 165,938 IB exams administered (International Baccalaureate, 2010), com-pared to 2,860,912 AP exams that same year.
32 The U.S. News AP criterion applies a 25% weight to the proportion of students completing an AP exam and a 75% weight to the proportion of students earning a score of 3 on at least one AP exam before or during their senior year (American Institutes for Research, 2013). For high schools mo-tivated to be ranked highly, this criterion could essentially encourage high schools to have as many students as possible sign up for AP exams and to sign up for many AP exams, in hopes that students will pass at least one.
33 Although AP programs have expanded considerably in recent decades, there is question as to whether this expansion has yielded equitable access. A College Board study (2013) indicated that Hispanic, African American,
35 American Secondary Education 43(2) Spring 2015
36
37 Growth & Achievement TrendsJudson & Hobson
38 and Native American students are significantly less likely to be enrolled in AP courses than White and Asian students, even when they have equivalent levels of readiness (College Board, 2013). Yet AP growth can at least partly be attributed to state and federal programs that have targeted low income students and aided the inclusion of underrepresented groups (Darity, Castel-lino, Tyson, Cobb, & McMillen, 2001; Klopfenstein, 2004). However, the persistent AP opportunity gap has led to suggestions for administrators to evaluate their AP programs, reconsider student expectations, offer more AP courses, and identify schools and districts where gaps have been narrowed or eliminated (The Education Trust, 2013). Research indicates that when at-tempts to recruit more underrepresented students into AP are successful, it remains difficult to reach parity because such efforts are often countered by the greater demand for AP in more affluent communities (Klugman, 2013).
39 The intent of this exploratory study was to take a step back and put into focus the tempo of AP growth. At the same time, an objective was to con-trast growth with a variable that is less well known: achievement over time. Therefore, research literature that focuses on the relationship of AP with col-lege matriculation, teacher practices, test reliability, and other related mat-ters, was not touched upon here. Instead, a stage has been set to scrutinize growth and question its alignment with achievement.
40 Data Sources
41 AP data were collected from the College Board that oversees AP exam de-velopment and administration. Databases were culled from various College Board publicly available and requested reports. The different types of data were not always available from the same periods of time. For example, al-though the number of AP exams and the scores on those exams were avail-able from 1992 through 2012, data regarding the number of AP students was available beginning in 1996. Data were commonly only available as aggre-gated categories, such as by ethnicity or grade level, but without the ability to create matrices that would permit the interaction of multiple variables. For example, although it was possible to examine achievement of 11th grade students or Hispanic students, data did not permit examination of 11th grade Hispanic students over time. Graduation rate data were gathered from the National Center for Education Statistics.
42 AP Growth
43 Overall Growth
44 To evaluate the change in overall AP participation over time, three variables were compared to one another: (a) number of high school graduates, (b) number of AP students, and (c) number of AP exams taken. High school
46 American Secondary Education 43(2) Spring 2015
47
48 Judson & HobsonGrowth & Achievement Trends
49 graduate numbers were compared to the number of AP students and the number of AP exams to determine the degree to which the growth of AP could simply be attributed to rising high school enrollment and increasing graduation success in American schools. AP students are defined by the Col-lege Board as students enrolled in a high school course designated as an AP preparatory course.
50 Data were available from the 17-year period of 1996 through 2012. Relationships among the variables were first assessed using the conserva-tive statistical test of Spearman correlation analyses. To evaluate if recent trends have affected covariance among the variables, the data were split into two groups: 1996-2004 (nine years) and 2005-2012 (eight years). The reason these two time periods were compared was to determine if what seems to be recent accelerated growth of AP truly is a new phenomenon or if the growth of AP has simply been on a steady upward march over an extended period of time. In other words, if the relationship among the three variables remained consistent during 1996-2004 and during 2005-2012, then we could say that there was a false perception that growth of AP had accelerated in most recent years. However, if the relationship, among the three variables, was different between the two time periods, that would indicate a more recent pattern change.
51 It is noted that the following AP exams were introduced since 1996. Analyses included these exams because the purpose was to evaluate the overall growth of AP and not specific exams. The addition of new AP exams over the years has been one obvious cause for growth, with these newer exams accounting for more than 15% of all AP exams administered in 2012.
52 • Chinese Language, introduced in 2007 (0.3% of 2012 exams)
53 • Environmental Science, introduced in 1998 (2.9% of 2012 exams)
54 • Human Geography, introduced in 2001 (2.7% of 2012 exams)
55 • Japanese Language, introduced in 2007 (0.1% of 2012 exams)
56 • Statistics, introduced in 1997 (4.2% of 2012 exams)
57 • World History, introduced in 2002 (5.7% of 2012 exams)
58 Data from 1996 through 2012 were evaluated to determine the relation-ships among the variables of high school graduates, quantity of AP students, and quantity of AP exams administered. All three of the variables were sig-nificantly correlated to one another for this seventeen year period (p < .01). Figure 1 reflects the changes from 1996 through 2012 among high school graduates, AP students, and AP exams administered.
60 American Secondary Education 43(2) Spring 2015
61
62 Growth & Achievement TrendsJudson & Hobson
63 Figure 1. High School Graduates, AP students, and AP exams, 1996-2012
|
|
64 4,000,000 |
|
65 Amount |
66 3,500,000 |
|
|
67 3,000,000 |
|
|
68 2,500,000 |
|
|
69 2,000,000 |
|
|
70 1,500,000 |
|
|
71 1,000,000 |
|
|
72 500,000 |
|
|
73 0 |
|
|
74 19961997199819992000200120022003200420052006200720082009201020112012 |
75
76
HS_Grads
77 AP_Students
78
79
AP_Exams
80 As indicated, data were split into two time period groups (1996-2004 and 2005-2012) and again correlations among the variables were examined. Analysis revealed that high school graduates, quantity of AP students, and quantity of AP exams administered, were all significantly correlated (p < .01) from 1996 through 2004. However, from 2005-2012, although the relation-ship between AP students and AP exams remained significant (p < .01), the rapid rate of increase of these two variables was out of step with the growth of high school graduates. During the most recent eight years, the increase in high school graduates was not significantly correlated to either the growth of AP students (p = .12, r =.59) or the increase in the number of AP exams (p = .12, r =.60).
81 In simple terms, although the number of high school graduates increased by 27.9% from 1996 through 2012, this was severely outpaced by the growth in the number of AP students and number of AP exams administered. During this time period, the quantity of AP students grew by 290.7% and the quantity of AP exams administered grew by 338.5 percent. This dissimilarity in growth rate has become even more acute in the past several years.
82 Growth by Grade Level
83 Grade specific data were available from 1997 through 2012. For each of these years, the number of AP exams and the number of AP students were greater for each successive grade, ninth- through twelfth-grade. That is, twelfth-grade always had, by far, the most AP test takers and the most AP exams and ninth-grade always had the fewest (Table 1).
85 American Secondary Education 43(2) Spring 2015
86
|
87 Judson & Hobson |
|
|
88 Growth & Achievement Trends |
|
|
89 Table 1. AP students in 1997 and 2012 |
|
|
||
|
|
|
|
|
|
|
|
90 Grade |
91 1997 |
92 2012 |
93 % Increase |
|
|
|
|
|
|
|
94 AP Students |
95 9 |
96 1,671 |
97 88,147 |
98 5175.1 |
|
|
99 10 |
100 41,487 |
101 364,913 |
102 779.6 |
|
|
103 11 |
104 219,427 |
105 781,437 |
106 256.1 |
|
|
107 12 |
108 309,169 |
109 824,869 |
110 166.8 |
111 Throughout the sixteen-year period, AP popularity was on the rise in all four grades. However, the growth rate was progressively greater for each preced-ing grade. Specifically, while the number of AP exams taken by twelfth-grade students grew a tremendous 211%, this was considerably less than the 353% growth in 11th grade or the 844% growth in tenth-grade. Even more dramatic was the 4952% growth in ninth-grade (Table 2).
112 Table 2. AP examinations in 1997 and 2012
113
|
|
114 Grade |
115 1997 |
116 2012 |
117 % Increase |
|
|
|
|
|
|
|
118 Exams taken |
119 9 |
120 1,829 |
121 92,407 |
122 4952.3% |
|
|
123 10 |
124 45,307 |
125 427,628 |
126 843.8 |
|
|
127 11 |
128 307,649 |
129 1,394,269 |
130 353.2 |
|
|
131 12 |
132 553,987 |
133 1,722,398 |
134 210.9 |
135 Growth by Ethnic Group
136 To evaluate how growth has been experienced among different ethnic groups, the AP test-taking data were compared to the total number of high school graduates, per subgroup. Reliable high school graduate data, per eth-nic group, were available only as recently as 2009 from the National Cen-ter for Education Statistics. The high school graduate values provide a sense of the magnitude of American high school students of each ethnicity, but are not strictly representative of enrollment because they typically represent twelfth-grade and obviously do not include those students who did not com-plete high school. By comparing the quantity of high school graduates to the number of AP exams per ethnic group, it was possible to assess the degree to which AP growth can be attributed to AP phenomena and not only to propor-tional enrollment growth of one ethnic group compared to others.
138 American Secondary Education 43(2) Spring 2015
139
140 Growth & Achievement TrendsJudson & Hobson
141 Although all ethnic groups experienced substantial growth in their num-ber of high school graduates, ranging from 25% growth among White stu-dents to 132% growth for Black students, these growth rates were dwarfed by the sheer explosion of AP exams taken by students of every ethnic group. The volume growth of AP exams ranged from a low of 191% among White students to 467% among Hispanic students. However, these growth percent-age results are skewed by the varying rates of subgroup student population growth during this time period. To address this issue, the ratio of AP exams to the number of graduates per subgroup provides a more sensible tactic for considering the growth of AP in light of the growth of student enrollment.
142 Table 3. AP Exams and High School Graduates, by Ethnic Group, 1997 and 2009
|
|
143 1997 |
|
|
|
144 2009 |
|
|
145 % increase |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
146 Ratio AP |
|
|
|
147 Ratio AP |
148 in ratio |
|
|
|
149 AP |
150 HS |
151 exams |
|
|
152 HS |
153 exams : |
154 AP exams : |
|
|
155 Ethnicity |
156 Exams |
157 Graduates |
158 graduates |
|
159 AP Exams |
160 Graduates |
161 graduates |
162 graduates |
|
|
|
|
|
|
|
|
|
|
|
|
|
163 Am. |
164 3,662 |
165 19,965 |
166 0.18 |
|
167 13,207 |
168 32,213 |
169 0.41 |
170 123.5% |
|
|
171 Indian |
|
|
|||||||
|
|
|
|
|
|
|
|
|
|
|
|
172 Asian |
173 117,650 |
174 90,222 |
175 1.30 |
|
176 396,202 |
177 163,575 |
178 2.42 |
179 85.8% |
|
|
180 Black |
181 34,514 |
182 207,684 |
183 0.17 |
|
184 186,083 |
185 481,698 |
186 0.39 |
187 132.5% |
|
|
188 Hispanic |
189 67,705 |
190 264,466 |
191 0.26 |
|
192 383,915 |
193 451,384 |
194 0.85 |
195 232.2% |
|
|
196 White |
197 584,533 |
198 1,504,174 |
199 0.39 |
|
200 1,699,161 |
201 1,883,382 |
202 0.90 |
203 132.2% |
|
|
|
|
|
|
|
|
|
|
|
|
204 As an example of how the exam-to-graduate ratio shifted over time, in 1997 for every Asian student graduating from high school, there were 1.3 AP ex-ams administered to Asian students from all grades; by 2009 there were 2.4 AP exams administered to Asian students for every Asian student graduating from high school. There was an upward shift among all ethnic groups. The ratio of AP exams administered to students of any one ethnic group, as com-pared to the number of high school graduates from that group, was greater in 2009 than in 1997.
205 The 85% growth of Asian students’ exam-to-graduate ratio was the least compared to the other groups; however, Asian students had substantially the highest ratio in both years. Hispanic students experienced the greatest shift in the exam-to-graduate ratio. In 1997, for every four AP exams administered to Hispanic students, only one Hispanic student was graduating high school. By 2009, the number of Hispanic students graduating high school was nearly equivalent to the number of AP exams administered to Hispanic students. Moreover, the exam-to-graduate ratio among Hispanic students was nearly equal to that of White students by 2009.
207 American Secondary Education 43(2) Spring 2015
208
209 Judson & HobsonGrowth & Achievement Trends
210 AP Achievement Trends
211 Overall Achievement Trend
212 Achievement results were available for the aggregate of all AP exams admin-istered from 1992 through 2012. Considering each year as the unit of analysis, regression analysis was applied to determine if a significant directional trend of mean scores existed across the years. The percent of students passing (i.e., obtaining a score of 3 or higher) on all AP exams decreased from the 1992 level of 65.5% to the 2012 level of 59.2% (Figure 2). Regression analysis revealed this was a significant downward trend (r-squared = 0.84, p < .001).
213 Figure 2. Percent of Students Passing (score 3 or above), All AP Exams
|
|
214 68 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
215 66 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
216 % |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
217 passing |
218 64 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
219 62 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
220 60 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
221 R² = 0.8414 |
|
|
|
|||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
222 58 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
223 56 |
|
|
|
|
|
|
|
224 2000 |
225 2001 |
226 2002 |
|
|
|
|
|
|
|
|
|
|
|
|
|
227 1992 |
228 1993 |
229 1994 |
230 1995 |
231 1996 |
232 1997 |
233 1998 |
234 1999 |
235 2003 |
236 2004 |
237 2005 |
238 2006 |
239 2007 |
240 2008 |
241 2009 |
242 2010 |
243 2011 |
244 2012 |
|
|||
245
246 Further, the individual achievement levels (i.e., scores of 1 through 5) were examined to evaluate trends over this 21-year period. Regression analy-ses were again applied to each achievement level to evaluate trends. Non-parametric correlation analyses were conducted to determine if the trend lines at the different levels of achievement varied in relationship to one an-other over time.
247 The regression analyses of the different achievement levels provided insight regarding the overall decline in the percent of students passing AP exams (Figure 3). For the highest score of 5, there was no discernible pattern over time (r-squared = .064, p = .27). There were, though, significant down-ward trends for scores of 4 (r-squared = .211, p = .04), scores of 3 (r-squared = .963, p < .001), and scores of 2 (r-squared = .699, p < .001). These down-ward trends were countered by a notable uptick in the percentage of students earning the lowest score of a 1 on an AP exam from 1992 through 2012 (r-squared = .897, p < .001).
249 American Secondary Education 43(2) Spring 2015
250
251 Growth & Achievement TrendsJudson & Hobson
252 Figure 3. Percent Achievement Levels, All AP Exams
253
|
254 30.0 |
|
|
|
|
|
|
255 % |
|
|
|
|
|
|
256 25.0 |
|
|
|
257 % 5s |
|
|
|
|
|
|
|
|
|
|
|
|
|
258 % 4s |
|
|
259 20.0 |
|
|
|
260 % 3s |
|
|
|
|
|
|
261 % 2s |
|
|
262 15.0 |
|
|
|
263 % 1s |
|
|
|
|
|
|
|
|
|
264 10.0 |
|
|
|
|
|
|
265 1992 |
266 1997 |
267 2002 |
268 2007 |
269 2012 |
|
270
271 There were significant negative correlations (p < .001) between the per-cent of students earning a score of 1 and the percentages of students earning a score of 2 (r = -.88), 3 (r = -.98), and 4 (r = -.88). An inverse and significant re-lationship (p < .001) was again found to exist when comparing the percentage of students passing (score of 3+) with those who earned a score of 1 (r = -.97).
272 Achievement by Grade Level
273 The ability to analyze achievement data, based on grade level, was hampered by the limited availability of data. Complete achievement data for 11th and 12th grades were available for the years 1997 through 2012. Achievement data for 9th and 10th grades were not available. However, the total num-ber of AP students enrolled in grades 9 and 10 were known. Additionally, the performance level of test-takers who were not in grades 11 or 12 was available. This category of not 11th or 12th grade included some students who were in grades below 9th grade, high school graduates, and what the College Board simply lists as “other.” For example, in 1997 there were a to-tal of 566,720 AP test-takers; of these, 517,297 were in 11th or 12th grade, 42,615 were in 9th or 10th grade, 189 were college students, and 6,619 were “other.” Beginning in 2005, the College Board began reporting the number of test-takers who were in grades below 9th (0.2% and 0.3% of all test-takers in 2005 and 2012, respectively).
274 Given these limitations, it was still possible to analyze student achieve-ment trends over time specifically for grades 11 and 12. The achievement over time for students not in grades 11 or 12 were also examined, but al-though those results can be attributed largely to students in 9th and 10th grade (86.2% of non-11th and non-12 grade test-takers in 1997 and 92.2% in 2012), the conglomeration of other student-test takers (e.g., less than ninth grade, college students) leaves this analysis questionable for its usefulness.
276 American Secondary Education 43(2) Spring 2015
277
278 Judson & HobsonGrowth & Achievement Trends
279 The AP exams that were completed by 11th grade students and earned a passing score of 3 or higher declined slightly from 62.5% in 1997 to 60.3% in 2012 (r-squared = 0.35, p < .05). The falloff in pass rates was more notice-able among 12th grade students. In 1997 65.2% of AP exams administered to 12th grade students received a score of 3 or higher, but by 2012 that rate had dipped to 58.2% (r-squared = 0.94, p < .001). Again, while it was desired to examine the achievement trends additionally for grades 9 and 10, the data only allowed for an examination of the achievement levels among students not enrolled in either 11th or 12th grades. The achievement trend for these non-11th and non-12th grade students also revealed a downward sloping regression line. In 1997 a passing score of 3 or better was achieved on 68.0% of the exams taken by non-11th and non-12th grade students; by 2012 that level had dipped conspicuously to 57.2% (r-squared = 0.90, p < .001).
280 Achievement by Ethnic Group
281 As indicated, data confirmed that AP test-taking has grown tremendously among all ethnic groups, and most remarkably among Hispanic students. Analyses of performance data, disaggregated by ethnicity, however, revealed differ-ent patterns that were at times converse to test-taking growth patterns. From 1997 through 2012, Asian and White students maintained relatively stable AP pass rates (scores of 3 or above) across all AP exams. In 1997, Asian students passed 67.0% of their AP exams and White students passed 65.5% of their AP exams. By 2012, these rates had little changed with 68.4% of AP exams administered to Asian students and 64.7% of AP exams administered to White students resulting in a score of a 3, 4, or 5. Across the 16 year period, there was little variation in the pass rates for these two groups, with pass rate stan-dard deviations being 1.4% for Asian students and 1.1% for White students.
282 Figure 4. Percent Passing (score 3 or above)
|
283 70.0% |
|
|
|
|
|
284 60.0% |
|
|
|
|
|
285 50.0% |
|
|
286 Asian |
|
|
|
|
287 White |
|
|
|
|
|
|
|
|
|
288 40.0% |
|
|
289 Hispanic |
|
|
|
|
290 Am. Indian |
|
|
|
|
|
|
|
|
|
291 30.0% |
|
|
292 Black |
|
|
|
|
|
|
|
|
293 20.0% |
|
|
|
|
|
294 1997 |
295 2002 |
296 2007 |
297 2012 |
|
298
300 American Secondary Education 43(2) Spring 2015
301
302 Growth & Achievement TrendsJudson & Hobson
303 Notable decreases in pass rates were apparent among the three ethnic groups typically considered underrepresented in higher education: American Indian, Black, and Hispanic. Figure 4 depicts the percentage of administered AP exams that resulted in a passing score per each ethnic group. Although all five groups experienced a small uptick in passing rates from 2010 to 2012, on the whole, data reveal that Asian and White students have sustained a pass rate percentage in the range of the mid-60s, while Black, Hispanic, and American Indian students have experienced substantial declines. American Indian students’ passing rate decreased from a high of 51.0% in 1997 to 45.5% in 2012 (r-squared = 0.56, p < .05). Black students maintained the lowest passing rate in all years and this decreased over time from 35.9% in 1997 to 29.1% in 2012 (r-squared = 0.81, p < .001). The sharpest rate of de-cline in passing rate occurred on exams administered to Hispanic students. In 1997, 61.1% of the exams administered to Hispanic students resulted in a passing score; however, in 2012, the passing rate for Hispanic students had dropped to 42.8% (r-squared = 0.94, p < .001).
304 Discussion
305 Overall Participation Growth
306 Analyses of the longitudinal AP datasets revealed findings that were aligned with common speculation, but also exposed some disturbing patterns. In summarizing, it is important to juxtapose the major findings of this study. As expected, it was revealed there has been rapid growth in AP test-taking. Dis-aggregation led to exposing where that growth has been most extensive. Yet, at the same time pass rates have dropped. In fact, among those groups where there have been the greatest increases in AP participation (i.e., grades 9 and 10, and Hispanic students), there exist the greatest declines in pass rates.
307 In this exploratory study, we are able to direct attention to patterns, but only provide reasonable conjecture for the cause. Each of these conjectures can be developed as hypotheses and further studied. The data examined in this study indicates the rampant growth of AP, but provides no indication of the mechanisms underpinning this growth. Logically, the cause for the growth of AP is speculated to be coming from multiple fronts. There are, of course, societal ethos promoting rigorous curriculum and urging the need for secondary schools to help students become college-ready. These are excel-lent goals and AP courses certainly seem to align well with these objectives.
308 An economist might view AP exams as a commodity and attempt to ex-plain the growth of AP as a supply and demand issue. Indeed there has likely been growth on the demand side for AP courses from parents and students. However, for this discussion, the focus is on two supply side mechanisms that warrant further study: (a) limited course choices for high ability students, and (b) in-class incentive recruitment practices.
310 American Secondary Education 43(2) Spring 2015
311
312 Judson & HobsonGrowth & Achievement Trends
313 As suggested, the push for public schools to compete for student enroll-ment, not just with public charter schools and private schools, but with other public schools, has become more commonplace. For example, themed mag-net schools commonly promote the availability of AP courses as a recruit-ment lure. There may then be pressure for schools to compete in their ability to provide students with AP courses. In so doing, many high schools have shifted course offerings that college-bound students often take. In particular, courses customarily labeled as advanced or honors now do double duty as AP preparation courses (Sieff, 2011). In many cases, students are limited in course selection to either an AP course or a regular course that is considered as being for students not on a college-track. In these cases, AP course enroll-ment increases because the supply of other options for high ability students has become limited.
314 Of course, just because a student enrolls in an AP course, there is no technical requirement to actually sign up for the associated AP exam. How-ever, students may encounter tremendous incentives and pressures to sign up for AP exams. This is the second supply side mechanism that needs to be studied further, test recruitment strategies. Many school districts and states (e.g. California, Missouri, Texas) provide subsidies to parents with financial need to help pay for AP exams. Financial aid for AP exams is a credible way to work toward leveling the playing field for needy students. However, an-other practice is less evenhanded. To entice students to complete AP exams, some high school AP teachers organize their courses in ways that discourage students from opting to not take the AP exam.
315 Anecdotes abound of AP teachers strongly encouraging students to attend AP study sessions after school or offering extra credit for attending such study sessions. Incentives can escalate to a point where students might consider their course grade at risk if they do not sign up for the exam. Specifi-cally, in some cases, teachers allow students to be exempt from taking a final course exam if they take the AP exam. It is difficult to determine how wide-spread this practice is, but the practice does exist. As a cursory examination, a Google search on the keywords of high school, syllabus, and AP was made. The first 25 syllabi that resulted from the search were examined. Of these, four specifically indicated that completing the AP exam was a requirement of the course or that it could be taken in lieu of a final exam or semester exam. A biased search that additionally included the keywords final exam and ex-empt led to dozens of high school syllabi revealing policies of trading taking an AP exam for exemption from a classroom exam. In these cases, students who are concerned about their grades might opt for taking the AP exam, even if they do not feel prepared to pass it. Even if the student does poorly on the AP exam, that score, which is reported in July by the College Board, will not affect their high school course grade.
316 It has been suggested here that an economist might look at AP growth as a result of supply and demand pressures. However, this simplistic view
318 American Secondary Education 43(2) Spring 2015
319
320 Growth & Achievement TrendsJudson & Hobson
321 assumes that AP has remained a consistent product over time. More likely, the stature of AP has evolved over the past twenty years and this has af-fected demand from multiple “customers.” It has moved from being a choice among very ambitious advanced students to being a staple of high school curricula. As suggested, the changed value of AP is embodied by a U.S. News and World Report annual generation of lists of the Best High Schools (Morse, 2013) for the United States and also per each state. Desire to be on these lists might unintentionally be contributing to not just the growth in AP test-taking revealed in this study, but also the decline in pass rates.
322 Overall Pass Rate Decline
323 Reasonable explanations that hypothesize the cause of AP growth, can also be listed as reasons for the decline in the AP pass rate. For example, it has been suggested that AP interest may be pressing schools to offer AP courses that are staffed with underprepared teachers (Driscoll, 2012). Additionally, there is evidence that school administrators have increased the offerings of AP courses to bolster the reputation of a school and to meet both real and perceived demands of the community (Iatarola, Conger, & Long, 2011).
324 It is also possible that there has been a shifting philosophy regarding the value of AP experience for students among administrators. In a survey of 200 administrators and 230 AP teachers (Burton, Whitman, Yepes-Baraya, Cline, & Myung-in Kim, 2002), responses indicated the administrators and teachers ranked the goal of students earning a grade of 3 or higher lowest among reasons the school had for students to take AP courses. Instead, the goals of experiencing college-level work, building confidence and interest in a subject, improving college admission chances, and gaining confidence in college success were all rated higher. If the goal of an AP course is then weighted more toward instilling confidence and promoting interest than to-ward passing the AP exam, it is not unreasonable to imagine the scope of recruitment to include students who may benefit from affective goals, but are academically underprepared. If these sentiments have shifted over time, then this could be one contributor to the marked increase in the proportion of students earning a score of 1 on AP exams.
325 Growth and Decline among Younger Students and Hispanic Students
326 Beyond overall patterns, this study revealed explosive test-taking growth among younger grades and among Hispanic students. The fact that AP growth has been relatively great among the younger students may be due to a view of AP as a stamp that signifies the course rigor that many believe should per-meate earlier grades.The increase in AP for grades 9 and 10 may simply be the result of schools tapping a larger available client pool. That is, if school administrators feel a need to grow their AP programs, a logical approach would be to offer new AP courses to earlier grades rather than to attempt to
328 American Secondary Education 43(2) Spring 2015
329
330 Judson & HobsonGrowth & Achievement Trends
331 entice advanced students in grades 11 and 12 who may already be burdened with AP courses.
332 At first glance, the growth in AP among Hispanic students might simply be attributed to the proportional growth in American high schools of His-panic students. In 1990, Hispanic students comprised 12% of all students enrolled in public schools and by 2010 comprised 23% (Aud et al., 2012). However, in comparing the AP growth among Hispanic students to high school graduation rates, this study revealed AP growth among Hispanic stu-dents to far exceed not just general student population growth, but to well outpace AP growth of other ethnic groups.
333 Contrasted to this growth in participation, the pattern of overall pass rates among Hispanics has been a marked downward trend. Yet this aggregate view of AP data is likely not disclosing the full story. Effects from individual AP exams contribute differently to overall results. While investigation of the contribution of individual AP exams warrants further and deeper study, for the purposes of this discussion, a general comparison of Hispanic students’ data was made between 2001 and 2012. The cursory examination revealed interesting findings. Although the overall pass rate for Hispanic students de-clined 9.7% from 2001 to 2012, among the 26 exams that were administered in both years, the decline in pass rate exceeded the average (i.e., was more negative) on only four exams: Art History, Music Theory, Spanish Literature, and Spanish Language. On 11 other exams, the pass rate for Hispanic stu-dents did decline from 2001 to 2012, but was still not as great as the nega-tive 9.7% average. On another 11 exams, the pass rate for Hispanic students increased from 2001 to 2012; this increase in pass rate for Hispanic students ranged from 0.6% on the Physics B exam to 21.0% on the French Language exam. However, among the exams completed in both 2001 and 2012 and by at least 10,000 Hispanic students in 2012, only 4 out of 15 had increasing pass rates from 2001 to 2012. There are notable contributors to the overall decreased Hispanic pass rates, such as the Spanish Language exam that had a pass rate of 93.9% in 2001 and 80.3% in 2012. However, the decreased pass rate value is more attributed to two other factors: (a) continued low and decreasing pass rates on high participation exams such as English Litera-ture & Composition (Hispanic pass rates were 37.5% in 2001 and 33.3% in 2012) and (b) newer high participation exams with relatively low pass rates, such as World History (Hispanic pass rate was 31.1% in 2012). This type of examination of individual AP subject exams and their relationship to ethnic group results is only scratched here and deserves greater study.
334 Declining Pass Rate
335 Commentary here is devoted to the declining pass rate observed over the past two decades. It would be an error of judgment to assume there is a correct pass rate and that we have moved away from some appropriate pass
337 American Secondary Education 43(2) Spring 2015
338
339 Growth & Achievement TrendsJudson & Hobson
340 rate over time. However, the decline has been fairly constant and should be considered. Some might say that the overall decline in pass rates of approxi-mately seven percentage points since 1992 is simply an artifact of widening the AP pool. This idea of a larger pool would lead to expected lower overall pass rates for two reasons. The first reason is that as more teachers and stu-dents were brought into the fold, preparation and prerequisite skills for both were less scrutinized. Second, over time, the goal of AP courses has possibly shifted to being less about getting a passing score on AP exams and more about gaining experience with college coursework. If this second cause is in play, then there would be less concern paid to achievement and more paid to experience.
341 This then calls attention to what is perhaps the most troubling statistic revealed in this study: the tremendous increase in the proportion of students earning a 1 on their AP exams. Several postulations have been offered re-garding why the number of AP test-takers has increased. These include re-cruitment strategies such as trading classroom test exemption for AP comple-tion, a drive to compete with other schools, and shifting philosophy toward the goal of AP courses. Yet, no matter, the causes behind AP test-taking, the overall decline in pass rate in 20 years masks the large uptick in the propor-tion of students who are, for lack of a better term, severely failing AP exams (i.e., earning a score of 1). This raises concern that an increasing number of underprepared students are just sitting for AP exams. This exploratory study has clearly opened up new questions to investigate. Chief among these is to evaluate why there has been this increase on the low-end of AP scores. For reasons mentioned, students may feel pressured to sign up for AP exams even if they are ill-prepared to pass. The widened recruitment pool seems to go hand-in-hand with the reduced pass rate, and nowhere are the consequences of growing AP more evident than in the data demonstrating the increased proportion of students severely failing AP exams.
342 Implications and Conclusions
343 The implications of this study can be interpreted through lenses held by re-searchers, practitioners, and policymakers. Several researchers have previ-ously questioned the merit of AP (e.g., Sadler, Sonnert, Tai, & Klopfenstein, 2010) and this line of inquiry should persist. Considering the results of this study pointing to growth that may be out of balance with achievement, more detailed research is still needed that examines the value of these courses and exams as well research into related school and classroom practices. In this vein, the “exchange rate” of AP credit should be evaluated. For example, if a student passes his or her AP Biology exam, does a university later exchange that passing score for a college credit that aligns with the student’s program of study? As pointed out earlier, additional investigation is also needed to
345 American Secondary Education 43(2) Spring 2015
346
347 Judson & HobsonGrowth & Achievement Trends
348 assess how schools recruit students into AP courses and encourage them to enroll for AP exams.
349 For practitioners and policymakers, these findings by themselves do not necessarily indicate a need to brake but they do point to the need to seriously consider how best to serve students. The mere existence of AP programs can-not be assumed to equate to excellence; instead alternatives such as honors courses must be considered. Policies that award schools status points for maintaining AP, IB, and Cambridge courses should be evaluated to assess if desired goals are genuinely being achieved. If indeed school administrators and teachers are looking to AP courses for the benefit of students gaining experience with college-level coursework, but with decreasing emphasis on passing the exams, then that assumption needs to also be questioned, i.e., just how many AP courses need a student enroll in and how many AP ex-ams does a student need to sit for before he or she has gained that desired “experience?”
350 Finally, it appears that the growing emphasis on the valuable goal of graduating college-ready high school students has often led to creating col-lege experiences in the high schools. What needs to be questioned here is if providing a college-level course is the same as readying a student for college. Just as a “junior” high school can attempt to enrich curriculum by introduc-ing high school curriculum in seventh- or eighth-grade, a high school can likewise attempt to be a “junior” university by providing college curriculum as early as ninth-grade. It is likely more important to take a step back and evaluate how to best take advantage of the high school environment in order to engage advanced students with rigorous curriculum, without attempting to mirror a university.
351 References
352 American Institutes for Research. (2013). Identifying top-performing high schools for the “best high schools” rankings: Analytical methodology and technical appendix. Re-trieved from http://static.usnews.com/documents/best-highschools/Identifying_Top_ Performing_High_Schools_April2013.pdf
353 Aud, S., Hussar, W., Johnson, F., Kena, G., Roth, E., Manning, E., Wang, X., and Zhang, J. (2012). The Condition of Education 2012 (NCES 2012-045). Washington, DC: U.S. De-partment of Education, National Center for Education Statistics. Retrieved from http:// nces.ed.gov/pubs2012/2012045.pdf
354 Burton, N. W., Whitman, N. B., Kim, R. M. I., Cline, F., & Yepes-Baraya, M. (2002). Minority student success: The role of teachers in Advanced Placement Program (AP) courses. Re-trieved from: http://onlinelibrary.wiley.com/doi/10.1002/j.2333-8504.2002.tb01884.x/ epdf
355 Challenge Success. (2013). The Advanced Placement program: Living up to its promise?
356 Retrieved from http://www.challengesuccess.org/research/white-papers.aspx
357 College Board. (2013). The 9th annual AP report to the nation. Retrieved from http://media. collegeboard.com/digitalServices/pdf/ap/rtn/9th-annual/9th-annual-ap-report-single-page.pdf
359 American Secondary Education 43(2) Spring 2015
360
361 Growth & Achievement TrendsJudson & Hobson
362 Darity, W., Castellino, D., Tyson, K., Cobb, C., McMillen, B. (2001, May). Increasing opportunity to learn via access to rigorous courses and programs: One strategy for closing the achievement gap for at risk and ethnic minority students. Raleigh, NC: State Board of Education. Retrieved from http://dpi.state.nc.us/docs/academicservices/ gifted/resources/reports/increasingopportunities.pdf
363 Driscoll, E. (2012, April 25). Why 50% of students fail AP exams and how to change that. Fox Business. Retrieved from http://www.foxbusiness.com/personal-fi-nance/2012/04/25/why-50-students-fail-ap-exams-and-how-to-change-that/
364 The Education Trust. (2013). Finding America’s missing AP and IB students. Retrieved from: http://edtrust.org/resource/finding-americas-missing-ap-and-ib-students/
365 Farkas, S., & Duffett, A. (2009). Growing pains in the advanced placement program: Do tough trade-offs lie ahead? Washington, DC: Thomas B. Fordham Foundation.
366 Iatarola, P., Conger, D., & Long, M. C. (2011). Determinants of high schools’ advanced course offerings. Educational Evaluation and Policy Analysis, 33(3), 340-359.
367 International Baccalaureate (2010). United States: A dynamic presence: Growth and char-acteristics of IB world schools. Retrieved from http://www.edtrust.org/missing_students Klopfenstein, K. (2004). The advanced placement expansion of the 1990s: How did tradi-
368 tionally underserved students fare? Education Policy Analysis Archives, 12(68), 1-15. Klopfenstein, K. (2010). Does the advanced placement program save taxpayers money?
369 The effect of AP participation on time to college graduation. In P.M. Sadler, G. Sonnert, R.H. Tai, & K. Klopfenstein, AP: A Critical Examination of the Advanced Placement Program (pp. 189–218). Cambridge, MA: Harvard Education Press.
370 Klugman, J. (2013). The advanced placement arms race and the reproduction of educa-tional inequality. Teachers College Record, 115(5), 1-34.
371 Koebler, J. (2012, January 11). Should high schools limit AP course enrollment. US News & World Report. Retrieved from http://www.usnews.com/education/blogs/high-school-notes/2012/01/11/should-high-schools-limit-ap-course-enrollment#comments Lichten, W. (2010). Whither advanced placement – now? In P.M. Sadler, G. Sonnert, R.H. Tai, and K. Klopfenstein, AP: A Critical Examination of the Advanced Placement Pro-
372 gram (pp. 233–243). Cambridge, MA: Harvard Education Press.
373 Morse, R. (2013). How U.S. News calculated the 2013 best high schools rankings. U.S. News and World Report. Retrieved from http://www.usnews.com/education/high-schools/articles/2013/04/22/how-us-news-calculated-the-2013-best-high-schools-rankings
374 Sadler, P. M. (2010). Key findings. In P.M. Sadler, G. Sonnert, R.H. Tai, and K. Klopfen-stein, AP: A Critical Examination of the Advanced Placement Program (pp. 233–243). Cambridge, MA: Harvard Education Press.
375 Sadler, P. M., Sonnert, G., Tai, R. H., & Klopfenstein, K. (Eds.) (2010). AP: A critical ex-amination of the Advanced Placement program. Cambridge, MA: Harvard Education Press.
376 Sieff, K. (2011, May 21). School districts move away from honors classes in favor of AP courses. The Washington Post. Retrieved from http://articles.washingtonpost. com/2011-05-21/local/35263428_1_honors-courses-honors-track-honors-students
378 Copyright of American Secondary Education is the property of American Secondary Education and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
General Document Comments 2
icons on the left to see existing comments.
0 archived comments