Moncrief yin school ratings housing factors final report

School Ratings and Housing Factors in Austin

An Independent Study Report

Christi J Moncrief Yin Department of Geography

Texas State University – San Marcos Geography / Business Administration

May 07, 2012

1 | P a g e C h r i s t i J . M o n c r i e f Y i n

Introduction

Under the direction of Dr. Brock Brown, Professor in the Texas State

University – San Marcos Geography Department and Dr. James LeSage,

Professor in the Texas State University – San Marcos Finance and Economics

Department, this research was conducted for the purpose of determining the

relationship between high school accountability ratings and housing factors

within the Austin MSA. Data was collected and organized into one of 62

groups based on location within the boundary zone of each public high

school building that graduates 12th graders in the Austin MSA.

Each high school building is issued an individual rating by the Texas

Education Agency, separate than that of the district in which it is located, so

this study will use the more precise, building-specific data, rather than the

broad district ratings. Housing factors such as median selling price, selling

price/ sq. foot, days on the market and school building’s percentage share of

total houses sold were examined. Annual school building accountability

ratings were collected from the Texas Education Agency and housing market

information was gathered from the Austin Board of REALTORS® (ABoR)

historical database.


Part One – Data Presentation

Annual school building accountability ratings were gathered for each of

the 62 high school buildings for each of the three years analyzed in this

study. The Texas Education Agency issues individual ratings to each building

separate from the rating issued to the district as a whole and those

individual ratings are the ones that this study looks at. The TEA ratings for

each building are based on student achievements on standardized tests,

building graduation rates, building attendance rates and various other

factors. Each building is then awarded a rating in one of the five following

categories, which are listed in descending order of achievement: Exemplary,

Recognized, Academically Acceptable, Academically Unacceptable, No Data /

Not Rated.

The Austin Board of REALTORS® maintains a historical database of

information pertaining to property sales over the past several years. Housing

factors can be cross-referenced by location within a particular school building

boundary and isolated to provide statistics on home sales within the

boundary of each building. This data can then be cross-referenced with date

and provide very specific data regarding sales statistics within a specific

timeframe and within a specific high school building boundary, as shown in

Figure 1.


Figure 1: GIS Representation of 62 High School Building Boundaries in the

Austin MSA


The data for this study comes from three separate annual periods.

August 1st through July 31st was selected as the annual period because this

correlates to the time period that new TEA ratings are released each year.

The logic behind this is that TEA ratings are released on the last day of July

each year and home buyer and sellers would use this information in making

buying and selling decisions from August 1st until the next year’s rating is

released on the last day of July. The most up to date data was used in this

study, representing three years of available data, released in July of 2008,

2009 and 2010. The three year periods are as follows:

2008 - 09: August 1st, 2008 – July 3st, 2009 (TEA Rating issued in July

2008)

2009 - 10: August 1st, 2009 – July 31st, 2010 (TEA Rating issued in

July 2009)

2010 - 11: August 1st, 2010 – July 31st, 2011 (TEA Rating issued in

July 2010)

A. Building Performance and Location

The raw data for 2008 – 09 is presented in Table 1. Only one school,

Westlake High School in Eanes Independent School District, achieved the

highest rating of “Exemplary” in 2008 - 09. Nine schools achieved a rating of


“Recognized” and forty schools achieved a rating of “Acceptable”. The large

number of “Academically Unacceptable” ratings is disturbing as this means

that these high schools did not sufficiently teach their students the required

information to pass the basic standardized test, did not graduate an ample

amount of their students and did not even see an acceptable amount of

attendance by their students during this year. Seven high schools in the

Austin MSA fit into this category during 2008 - 09, which is a full 11% of the

high schools. There were also five schools that received a “No Data” rating

during 2008 - 09 which means that the TEA did not publish a rating for this

building. The likely reason for this is that the building is new and has not

been in existence long enough to meet the state requirements to be

awarded a rating.


District Name Building Name

TEA Rating - Individual Building

Total Units Sold

Median Median Median Median Median Median Median Median

Beds Baths SqFt Listing Price LP/SqFt

Selling Price SP/SqFt DOM

Austin_ISD Akins_HS Acceptable 614 3 2 1,620 $159,900 $101.35 $158,000 $99.14 34

Austin_ISD Anderson_HS Acceptable 750 3 2 1,689 $225,000 $145.01 $220,000 $140.10 34

Austin_ISD Austin_HS Acceptable 841 3 2 1,698 $309,000 $218.06 $295,000 $208.49 52

Austin_ISD Bowie_HS Recognized 750 4 2 2,310 $260,000 $116.98 $256,250 $113.71 39

Austin_ISD Crockett_HS Unacceptable 531 3 2 1,441 $169,900 $119.54 $168,000 $116.52 30

Austin_ISD Eastside_Memorial NoData 111 3 1 1,140 $157,900 $135.14 $155,000 $130.67 55

Austin_ISD Lanier_HS Acceptable 224 3 2 1,364 $142,968 $99.98 $138,500 $97.97 38

Austin_ISD LBJ_HS Acceptable 100 3 2 1,372 $123,500 $86.60 $115,500 $86.47 46

Austin_ISD McCallum_HS Acceptable 779 3 2 1,334 $278,950 $206.58 $269,000 $200.58 42

Austin_ISD Reagan_HS Unacceptable 239 3 2 1,498 $149,950 $102.64 $147,000 $100.71 47

Austin_ISD Travis_HS Acceptable 363 2 1 1,115 $189,000 $183.21 $177,500 $177.34 38

Bartlett_ISD Bartlett_HS Acceptable 1 3 2 3,862 $129,000 $33.40 $132,150 $34.22 78

Bastrop_ISD Bastrop_HS Unacceptable 325 3 2 1,722 $129,750 $77.57 $124,900 $75.47 47

Bastrop_ISD Cedar_Creek_HS NoData 0 0 0 0 $0 $0.00 $0 $0.00 0

Blanco_ISD Blanco_HS Acceptable 8 3 2 1,500 $190,750 $128.38 $173,500 $119.07 90

Burnet_ISD Burnet_HS Acceptable 97 3 2 1,475 $110,900 $73 $97,000 $70.38 67

Comal_ISD Canyon_HS Acceptable 6 3 2 1,720 $193,500 $109.27 $188,000 $105.83 78

Comal_ISD Canyon_Lake_HS Acceptable 16 3 2 1,779 $207,400 $108.12 $191,250 $102.75 107

Comal_ISD Smithson_Valley Acceptable 4 3 2 1,547 $151,950 $87.53 $134,060 $82.42 111

Del _Valle_ISD Del_Valle_HS Acceptable 260 3 2 1,562 $103,500 $65.21 $101,950 $62.87 38

Dripping_Springs Dripping_Springs Recognized 194 4 2 2,458 $307,300 $124.12 $295,000 $119.41 64

Eanes_ISD Westlake_HS Exemplary 345 4 3 2,874 $522,000 $193.99 $507,500 $186.59 50

Elgin_ISD Elgin_HS Unacceptable 154 3 2 1,698 $99,900 $60.37 $98,700 $60.32 41

Florence_ISD Florence_HS Acceptable 27 3 2 1,376 $119,900 $68.80 $114,900 $68.41 40

Georgetown_ISD East_View_HS Acceptable 515 3 2 1,958 $180,000 $96.76 $177,500 $94.26 55

Georgetown_ISD Georgetown_HS Acceptable 25 3 2 1,792 $145,000 $88.03 $143,000 $85 106

Gonzales_ISD Gonzales_HS Acceptable 8 3 2 2,104 $181,923 $95.46 $168,650 $88 70

Granger_ISD Granger_HS Acceptable 11 3 2 1,330 $89,750 $66.85 $87,000 $68.95 69

Hays_ISD Jack_C_Hays_HS Acceptable 330 3 2 1,903 $160,000 $90.84 $158,350 $89.25 46

Hays_ISD Lehman_HS Acceptable 288 3 2 1,817 $134,900 $73.73 $132,700 $71.98 42

Hutto_ISD Hutto_HS Acceptable 289 3 2 1,858 $125,000 $71.47 $125,000 $69.55 41

Jarrell_ISD Jarrell_HS Recognized 126 2 2 1,619 $179,900 $109.28 $169,250 $103.31 74

Jonhnson_City Lyndon_B_Johnson Acceptable 17 3 2 1,728 $209,900 $111.78 $199,900 $110.53 84

Lago_Vista_ISD Lago_Vista_HS Recognized 162 3 2 1,657 $179,250 $108.44 $174,750 $105.09 56

Lake_Travis_ISD Lake_Travis_HS Acceptable 489 3 2 2,582 $339,000 $134.86 $325,000 $129.04 68

Leander_ISD Cedar_Park_HS Recognized 644 4 2 2,681 $270,000 $108.16 $264,995 $104.63 62

Leander_ISD Leander_HS Acceptable 533 3 2 1,739 $144,900 $84.42 $143,000 $84 41

Leander_ISD Rouse_HS NoData 19 3 2 1,788 $159,900 $86.46 $153,500 $85.49 30

Leander_ISD Vandegrift_HS NoData 1 3 2 2,484 $289,900 $116.71 $287,000 $115.54 39

Leander_ISD Vista_Ridge_HS Acceptable 567 3 2 2,050 $179,900 $89.18 $178,000 $87.74 46

Lexington_ISD Lexington_HS Acceptable 18 3 2 1,784 $110,950 $64.06 $107,397 $60.15 34

Liberty_Hill_ISD Liberty_Hill_HS Acceptable 105 3 2 2,039 $214,777 $107.10 $210,000 $104.17 71

Lockhart_ISD Lockhart_HS Acceptable 114 3 2 1,518 $102,950 $69.25 $96,925 $68.12 79

Luling_ISD Luling_HS Acceptable 26 3 2 1,379 $86,000 $65.14 $80,875 $58.78 55

Manor_ISD Manor_HS Unacceptable 313 3 2 1,675 $117,900 $69.11 $115,000 $67.65 37

Marble_Falls_ISD Marble_Falls_HS Acceptable 107 3 2 2,015 $249,000 $113.00 $232,000 $104.00 90

Pflugerville_ISD Hendrickson_HS Acceptable 321 4 2 2,236 $169,900 $78.99 $166,500 $77.71 41

Pflugerville_ISD John_B_Connally Acceptable 420 3 2 1,550 $145,000 $98.66 $143,250 $97.05 40

Pflugerville_ISD Pflugerville_HS Acceptable 457 3 2 1,824 $146,995 $81.76 $145,000 $80.51 38

Prairie_Lea_ISD Prairie_Lea_ISD Acceptable 1 4 1 4,514 $159,500 $35.33 $159,900 $35.42 245

Round_Rock_ISD Cedar_Ridge_HS NoData 1 3 2 1,654 $177,900 $107.56 $175,000 $105.80 81

Round_Rock_ISD McNeil_HS Acceptable 656 3 2 1,960 $189,900 $99.07 $185,500 $97.48 34

Round_Rock_ISD Round_Rock_HS Acceptable 494 3 2 2,184 $191,450 $91.25 $189,950 $89.94 43

Round_Rock_ISD Stony_Point_HS Acceptable 859 3 2 1,945 $154,900 $83.90 $152,000 $82.72 46

Round_Rock_ISD Westwood_HS Recognized 463 4 2 2,149 $269,900 $117.19 $260,000 $114.41 35

San_Marcos_ISD San_Marcos_HS Acceptable 138 3 2 1,586 $141,700 $94.43 $136,500 $91.51 46

Smithville_ISD Smithville_HS Acceptable 91 3 2 1,575 $115,000 $74.66 $99,900 $67.71 44

Taylor_ISD Taylor_HS Unacceptable 138 3 2 1,476 $107,000 $70.47 $105,000 $67.43 37

Thorndale_ISD Thorndale_HS Recognized 12 3 2 1,476 $107,000 $70.47 $105,000 $67.43 37

Thrall_ISD Thrall_HS Recognized 15 3 2 1,633 $150,000 $84.94 $143,900 $84.94 102

Waelder_ISD Waelder_HS Unacceptable 0 0 0 0 0 0 0 0 0

Wimberley_ISD Wimberley_HS Recognized 151 3 2 1,756 $180,000 $107.76 $175,000 $102.71 63

Table 1: 2008 - 09 Complete Data Table


Figure 2 illustrates the spatial distribution of high school accountability

ratings during 2008 - 09. Take note that only one high school building

received the highest rating of “Exemplary”.

Figure 2: 2008 - 09 High School TEA Ratings in the Austin MSA


Data for 2009 - 10 is presented in Table 2. Four schools joined

Westlake High School this year in achieving the highest rating of

“Exemplary” for a total of five schools to earn this honor. Fifteen schools

achieved a rating of “Recognized” and thirty-three schools achieved a rating

of “Acceptable”. The number of “Academically Unacceptable” ratings fell

slightly this year, with a total of only five buildings receiving this rating. It is

noteworthy that of the seven buildings receiving the “Academically

Unacceptable” rating during 2008 - 09, all but two, Reagan High School in

Austin ISD and Manor High School in Manor ISD, improved their standing for

this year’s ratings. The other three schools in this category this year are

buildings that fell from a higher rating and are new to this group. Three

buildings received a “No Data” rating during 2009 - 10.




Total Units Sold







Austin_ISD Bowie_HS Acceptable 842 4 2 2,344 $270,000 $117.27 $268,500 $114.55 29

Austin_ISD Crockett_HS Acceptable 600 3 2 1,426 $169,700 $122.22 $166,000 $121.51 22

Austin_ISD Eastside_Memorial Unacceptable 96 3 1 1,123 $170,000 $147.02 $167,500 $144.91 63


Austin_ISD LBJ_HS Unacceptable 129 3 2 1,353 $129,900 $95.77 $127,000 $91.72 50


Austin_ISD Reagan_HS Unacceptable 290 3 2 1,366 $154,700 $109.73 $152,220 $106.83 36



Bastrop_ISD Bastrop_HS Acceptable 336 3 2 1,716 $125,000 $76.26 $122,500 $73.13 39

Bastrop_ISD Cedar_Creek_HS NoData 3 4 2 2,432 $100,000 $36.37 $109,000 $41.29 16


Burnet_ISD Burnet_HS Acceptable 90 3 2 1,607 $124,825 $78 $121,550 $75.43 61

Comal_ISD Canyon_HS Acceptable 6 3 2 2,358 $282,344 $122.91 $278,250 $113.83 120

Comal_ISD Canyon_Lake_HS Recognized 20 3 2 1,754 $187,400 $110.55 $176,750 $104.73 87

Comal_ISD Smithson_Valley Acceptable 4 3 3 2,165 $247,800 $100.51 $240,500 $96.17 208

Del _Valle_ISD Del_Valle_HS Recognized 340 3 2 1,469 $100,000 $67.99 $99,250 $66.96 32

Dripping_Springs Dripping_Springs Recognized 237 4 2 2,706 $310,000 $116.22 $300,000 $113.17 57


Elgin_ISD Elgin_HS Acceptable 182 3 2 1,715 $100,000 $60.96 $99,450 $60.15 46

Florence_ISD Florence_HS Recognized 22 3 2 1,981 $136,800 $78.58 $137,500 $79.50 77

Georgetown_ISD East_View_HS Recognized 627 3 2 1,929 $179,900 $97.48 $173,500 $95.35 54


Gonzales_ISD Gonzales_HS Acceptable 7 3 2 1,876 $189,000 $80.94 $140,000 $77 100




Hutto_ISD Hutto_HS Recognized 366 3 2 1,845 $126,000 $71.07 $125,000 $70.94 35

Jarrell_ISD Jarrell_HS Exemplary 158 2 2 1,596 $158,450 $103.69 $152,000 $99.57 68


Lago_Vista_ISD Lago_Vista_HS Exemplary 209 3 2 1,780 $189,900 $106.38 $184,900 $102.78 79

Lake_Travis_ISD Lake_Travis_HS Acceptable 695 3 2 2,586 $330,000 $130.73 $316,500 $125.74 60


Leander_ISD Leander_HS Acceptable 635 3 2 1,809 $145,000 $82.93 $144,800 $82 39

Leander_ISD Rouse_HS Recognized 91 3 2 1,994 $156,000 $84.37 $154,500 $84.69 31

Leander_ISD Vandegrift_HS NoData 131 4 3 3,340 $398,000 $122.42 $388,500 $119.98 42

Leander_ISD Vista_Ridge_HS Recognized 600 3 2 2,119 $189,690 $91.63 $188,000 $90.21 41

Lexington_ISD Lexington_HS Acceptable 33 3 2 1,832 $149,000 $87.06 $147,000 $78.78 62

Liberty_Hill_ISD Liberty_Hill_HS Recognized 101 3 2 2,223 $234,900 $113.19 $225,000 $108.85 60

Lockhart_ISD Lockhart_HS Recognized 130 3 2 1,512 $106,900 $70.23 $99,750 $66.73 82

Luling_ISD Luling_HS Unacceptable 18 3 2 1,334 $79,700 $58.59 $76,825 $56.04 65

Manor_ISD Manor_HS Unacceptable 354 3 2 1,734 $124,995 $68.00 $123,450 $67.48 28

Marble_Falls_ISD Marble_Falls_HS Recognized 122 3 2 1,959 $224,950 $106.09 $215,000 $102.99 77

Pflugerville_ISD Hendrickson_HS Recognized 359 4 2 2,277 $170,900 $78.19 $167,000 $76.52 40


Pflugerville_ISD Pflugerville_HS Recognized 478 3 2 1,822 $147,500 $83.26 $144,975 $82.11 33



Round_Rock_ISD McNeil_HS Recognized 701 3 2 1,981 $194,900 $102.21 $192,000 $100.27 31

Round_Rock_ISD Round_Rock_HS Acceptable 577 3 2 2,145 $195,000 $92.09 $190,000 $90.69 34

Round_Rock_ISD Stony_Point_HS Acceptable 931 3 2 1,990 $155,000 $82.95 $154,900 $81.83 42

Round_Rock_ISD Westwood_HS Exemplary 563 3 2 2,099 $263,200 $113.72 $258,000 $109.66 28


Smithville_ISD Smithville_HS Acceptable 83 3 2 1,780 $129,000 $77.86 $124,000 $73.53 57

Taylor_ISD Taylor_HS Acceptable 149 3 2 1,428 $93,000 $61.71 $91,000 $60.77 47

Thorndale_ISD Thorndale_HS Exemplary 14 3 2 1,800 $77,950 $70.57 $75,500 $64.16 86

Thrall_ISD Thrall_HS Acceptable 21 3 2 1,513 $114,900 $73.98 $110,900 $73.98 28

Waelder_ISD Waelder_HS Acceptable 3 3 2 1536 250000 119.55 250000 118.32 130

Wimberley_ISD Wimberley_HS Exemplary 193 3 2 1,780 $199,900 $112.19 $196,250 $109.38 62



Figure 3 shows the spatial distribution of high school accountability

ratings during 2009 – 10, and illustrates the clustering of four out of the five

“Academically Unacceptable” schools in central Austin.

Figure 3: 2009 -10 High School TEA Ratings in the Austin MSA


Data for 2010 - 11 is presented in Table 3. Many buildings have

achieved the highest rating of “Exemplary” this year, seven in all, more than

any of the other study years. Twenty-four schools achieved a rating of

“Recognized” and twenty-eight schools achieved a rating of “Acceptable”.

The number of “Academically Unacceptable” ratings fell to one this year,

with Eastside Memorial High School in Austin ISD being the sole building that

could not improve its standing. Two buildings received a “No Data” rating

during the final year of study.




Total Units Sold







Austin_ISD Bowie_HS Recognized 728 4 2 2,387 $275,000 $115.86 $271,000 $113.24 42

Austin_ISD Crockett_HS Acceptable 500 3 2 1,429 $163,000 $117.01 $159,000 $113.47 43

Austin_ISD Eastside_Memorial Unacceptable 83 3 1 1,160 $165,000 $136.49 $161,000 $130.89 47


Austin_ISD LBJ_HS Acceptable 107 3 2 1,336 $113,000 $81.58 $110,000 $76.99 50


Austin_ISD Reagan_HS Acceptable 258 3 2 1,504 $146,000 $96.07 $139,850 $92.67 49



Bastrop_ISD Bastrop_HS Acceptable 271 3 2 1,705 $124,900 $72.39 $123,500 $68.80 60

Bastrop_ISD Cedar_Creek_HS NoData 65 3 2 1,840 $129,500 $71.09 $125,000 $69.38 27


Burnet_ISD Burnet_HS Recognized 100 3 2 1,793 $166,700 $88 $153,000 $82.90 84

Comal_ISD Canyon_HS Recognized 13 3 2 1,702 $169,000 $99.61 $166,700 $97.87 111

Comal_ISD Canyon_Lake_HS Recognized 20 3 2 1,838 $177,200 $99.14 $172,250 $93.27 108

Comal_ISD Smithson_Valley Recognized 6 3 2 2,151 $244,900 $98.56 $236,800 $92.04 65

Del _Valle_ISD Del_Valle_HS Recognized 307 3 2 1,587 $89,600 $57.48 $85,250 $55.61 51

Dripping_Springs Dripping_Springs Exemplary 251 4 3 2,820 $325,000 $118.58 $319,000 $114.42 73


Elgin_ISD Elgin_HS Acceptable 142 3 2 1,866 $94,950 $52.27 $92,700 $50.23 53

Florence_ISD Florence_HS Acceptable 18 3 2 1,833 $111,995 $58.34 $108,500 $57.35 59

Georgetown_ISD East_View_HS Recognized 607 3 2 2,045 $179,900 $93.78 $174,900 $91.31 59


Gonzales_ISD Gonzales_HS Recognized 6 3 2 1,405 $82,900 $72.28 $76,850 $67 110




Hutto_ISD Hutto_HS Recognized 305 3 2 1,877 $114,975 $62.14 $112,000 $60.23 46

Jarrell_ISD Jarrell_HS Recognized 144 3 2 1,597 $158,475 $104.91 $152,250 $101.20 71


Lago_Vista_ISD Lago_Vista_HS Recognized 196 3 2 1,794 $199,450 $102.93 $187,750 $98.82 74

Lake_Travis_ISD Lake_Travis_HS Exemplary 693 4 3 2,721 $354,900 $134.41 $347,500 $130.19 67


Leander_ISD Leander_HS Recognized 573 3 2 1,844 $144,900 $78.97 $143,475 $77 48

Leander_ISD Rouse_HS Recognized 108 3 2 2,026 $149,900 $80.39 $146,750 $78.27 46

Leander_ISD Vandegrift_HS Exemplary 230 4 3 3,252 $399,500 $131.39 $392,500 $126.14 47

Leander_ISD Vista_Ridge_HS Recognized 475 3 2 2,175 $189,900 $89.12 $186,500 $87.41 51

Lexington_ISD Lexington_HS Exemplary 10 3 2 1,679 $85,925 $68.24 $78,750 $64.03 88

Liberty_Hill_ISD Liberty_Hill_HS Recognized 120 3 2 2,444 $269,950 $106.38 $266,250 $102.64 68

Lockhart_ISD Lockhart_HS Recognized 116 3 2 1,567 $105,000 $69.38 $102,100 $63.80 76

Luling_ISD Luling_HS Acceptable 8 3 2 1,666 $74,950 $43.61 $71,005 $42.46 66

Manor_ISD Manor_HS Acceptable 344 3 2 1,727 $98,500 $59.74 $93,500 $57.77 55

Marble_Falls_ISD Marble_Falls_HS Acceptable 127 3 2 2,322 $315,000 $125.11 $305,000 $121.11 94

Pflugerville_ISD Hendrickson_HS Recognized 317 4 2 2,277 $164,500 $73.47 $161,900 $72.79 56


Pflugerville_ISD Pflugerville_HS Recognized 357 3 2 1,868 $139,900 $73.36 $137,900 $71.69 48



Round_Rock_ISD McNeil_HS Recognized 521 3 2 2,078 $196,690 $97.95 $191,500 $95.63 52

Round_Rock_ISD Round_Rock_HS Recognized 581 4 2 2,274 $194,000 $87.48 $189,900 $86.10 52

Round_Rock_ISD Stony_Point_HS Recognized 634 3 2 2,035 $149,900 $77.44 $147,575 $75.37 53

Round_Rock_ISD Westwood_HS Exemplary 513 4 2 2,210 $269,800 $118.42 $261,000 $114.09 35


Smithville_ISD Smithville_HS Recognized 69 3 2 1,780 $140,000 $78.95 $140,000 $74.55 95

Taylor_ISD Taylor_HS Acceptable 127 3 2 1,444 $87,000 $58.87 $79,900 $55.06 66

Thorndale_ISD Thorndale_HS Acceptable 12 3 2 2,141 $182,400 $80.19 $168,000 $75.61 88

Thrall_ISD Thrall_HS Exemplary 15 3 2 1,467 $80,000 $70.75 $70,000 $67.71 55

Waelder_ISD Waelder_HS Acceptable 0 0 0 0 0 0 0 0 0

Wimberley_ISD Wimberley_HS Recognized 184 3 2 1,930 $212,250 $110.60 $199,250 $106.48 84



Figure 4 shows the spatial distribution of high school accountability

ratings during 2010 – 11, where only one high school building received the

lowest rating of “Academically Unacceptable”.

Figure 4: 2010 - 11 High School TEA Ratings in the Austin MSA


B. Descriptive Statistics for Housing Factors

Summary statistics for all three years will provide a baseline for

analyzing the data in the growth regressions for the second part of this

report. The means, medians, standard deviations, minimums and maximums

are provided for all of the housing factor variables taken into account in this

study and shown individually for each of the three years being analyzed. The

following tables provide summary statistics for the days on the market, units

sold, median number of bedrooms, median number of bathrooms, sales

price in thousands and building ratings. In order to summarize the data

numerically, each TEA rating was assigned a number, as follows:

Exemplary: 4

Recognized: 3

Academically Acceptable: 2

Academically Unacceptable: 1

No Data: 0

The summary statistics for 2008 - 09 are presented in Table 4. Note

that the mean exceeds the median for the variables of days on market, units

sold and price in thousands indicating a right-skewed distribution of these

variables. The beds, baths and building ratings variables have means and

medians that are almost identical in this year indicating symmetry in those

housing factor variables.


Variables mean median standard dev. min max

days on market 56.3387 46.0000 33.4600 0.0000 245.0000

units sold 252.6290 152.5000 253.7957 0.0000 859.0000

beds 2.9839 3.0000 0.6651 0.0000 4.0000

baths 1.9032 2.0000 0.4327 0.0000 3.0000

price in thous. 166626.6452 156500.0000 77023.6907 0.0000 507500.0000

building ratings 1.9032 2.0000 0.8039 0.0000 4.0000

Table 4: 2008 - 09 Summary Statistics

Table 5 presents summary statistics for 2009 - 10. In a similar case as

2008 - 09, the means are larger than the medians for the days on market,

units sold and price in thousands variables indicating a right-skewed

distribution. The building ratings this year have a mean that is more than

10% larger than the median as well, demonstrating a right skew in this

variable. The beds and baths are once again showing symmetry.

Variables mean median standard dev. min max

days on market 54.0968 45.5000 31.6967 16.0000 208.0000

units sold 293.4516 187.5000 286.4301 3.0000 1124.0000

beds 3.1129 3.0000 0.4474 2.0000 4.0000

baths 2.0323 2.0000 0.3119 1.0000 3.0000

price in thous. 177204.2742 157500.0000 81803.1098 44000.0000 522500.0000




Table 6 presents summary statistics for 2010 - 11. As in the first two

years there is a right-skew in the days on the market, units sold and price in

thousands variables. And in a story similar to 2008 - 09, the other three

variables, beds, baths and building ratings, depict symmetry. Also note that

the standard deviation for the price in thousands variable is much farther

away from zero in 2010 - 11 than it is in 2008 - 09 or 2009 - 10. This larger

standard deviation indicates a wider spread in the distribution of this

variable across the 62 high school building boundaries reflecting a greater

inequality than in the previous two years. This means that there are

significant numbers of properties selling in the higher price brackets and

significant numbers of properties selling in the lower price brackets but less

properties selling in the mid-range price brackets.

variables mean median standard dev. min max

days on market 62.8387 54.0000 26.4644 0.0000 183.0000

units sold 268.4355 190.0000 263.1458 0.0000 1247.0000

beds 3.0968 3.0000 0.5642 0.0000 4.0000

baths 2.0000 2.0000 0.4049 0.0000 3.0000

price in thous. 172412.5806 151625.0000 92970.3075 0.0000 595000.0000



Figures 5, 6 and 7 show the median selling price of homes for all three

years. Homes located within the boundary of an exemplary rated high school


sold at higher prices than homes located within the boundary of lower rated

schools in all three years.

Figure 5: 2008 - 09 Median Selling Price of Homes in the Austin MSA

Grouped by High School Accountability Rating

Series1 $-

$200,000.00

$400,000.00

$600,000.00






Series1 $-

$50,000.00

$100,000.00

$150,000.00

$200,000.00

$250,000.00

Series1 $-

$100,000.00

$200,000.00

$300,000.00


There is one more noteworthy relationship which is consistent among

all three years of collected data. The schools with the largest quantity of

housing units sold per year are not the schools which are rated the highest.

In fact, the majority of schools with large quantities of housing units sold are

rated at merely “Academically Acceptable”. Figures 8, 9 and 10 illustrate the

top ten schools for each year with the largest quantity of housing units sold.

Figure 8: Top Ten Schools with the Largest Quantity of Housing Units Sold in

the Austin MSA in 2008 - 09, Grouped by School Accountability Rating

Acceptable

Recognized

Exemplary

Unacceptable


Figure 9: Top Ten Schools with the Largest Quantity of Housing Units Sold in

the Austin MSA in 2009 - 10, Grouped by School Accountability Rating

Figure 10: Top Ten Schools with the Largest Quantity of Housing Units Sold

in the Austin MSA in 2010 - 11, Grouped by School Accountability Rating

Acceptable

Recognized

Exemplary

Unacceptable

Acceptable

Recognized

Exemplary

Unacceptable


C. Relationship between Housing Factors

A correlation coefficient is used to summarize the strength of the linear

association between the variables. If the variables go up and down together,

the correlation coefficient will be positive. If the variables go up and down in

opposition, the correlation coefficient will be negative. Keep in mind however

that correlation is not causation. It would be easy to misinterpret the

correlation coefficient tables and naively arrive at a false conclusion that

building ratings have an impact on housing factors. Part two of this study

will analyze in detail the connection between these factors to determine

whether or not there may be a causal relationship.

Table 7 presents correlation coefficients for 2008 - 09. It is clear by

the data in the building ratings column that there is a positive relationship

between building ratings and the other housing factor variables, especially

the median price.

Variables price/sf days on market median price building ratings

price/sf 1.0000 -0.0241 0.7887 0.2631

days on market 1.0000 0.1301 0.1957

median price 1.0000 0.4128

building ratings 1.0000

Table 7: 2008 - 09 Correlation Coefficients


Correlation coefficients for 2009 - 10 are exhibited in Table 8. Positive

correlations between building ratings and the other variables are also

present in this year, although significantly less than in 2008 – 09, perhaps

due to the unique national housing market trends during this time period.


price/sf 1.0000 0.0756 0.7715 0.0969

dom 1.0000 0.1924 0.1511




Table 9 presents correlation coefficients for 2010 - 11. Once again the

building ratings column displays positive relationships between building

ratings and all other housing factor variables.


price/sf 1.0000 0.0204 0.8036 0.1124

days on market 1.0000 -0.0380 0.1206





In these correlations, median price and price per square foot are

strongly related suggesting that either one is a good measure of general

price trends. Otherwise, the housing factors are not strongly interrelated and

are suitable for robust statistical analysis.


Part Two – Data Analysis

Spatial dependence implies that observations (sales price of homes,

for example) depend on other observations (building ratings of the local high

school, for example) either at that location or at a location in relative

proximity to the location of the original observation.

Growth regressions are an important statistical analysis tool as they

can determine causal relationships between observations while taking into

account outside variables and explain the interdependent relationship of one

observation to another. Nothing is independent of anything else but

sometimes it’s difficult to understand the connectivity of relationships

without a close analysis. In this section multiple regression models will be

used to test the role of school quality (and other control variables) verses

three key housing indicators:

1) Total Price

2) Days on the Market

3) Price per Square Foot

A. Total Price

The first three growth regressions presented use “Price in Thousands”

as a dependent variable. These three regressions examine how building


ratings affect the selling price of homes within their own boundary as well as

homes within the neighboring boundaries. “Neighboring boundaries” refer to

the six geographically closest high school zones to the zone of the original

observation. The impact that is shown in the neighboring district is the total

impact among all of the districts; therefore it must be divided by six to

determine the impact on one individual neighboring district.

Figure 11 shows the growth regression for 2008 - 09 using “Price in

Thousands” as the dependent variable. The R-squared value of the model

indicates a low value for prediction overall, but the building ratings are

reporting a substantial influence on median price which is statistically

significant at the 70% level. What this means is that for each single increase

in building rating (i.e. a jump from Academically Acceptable to Recognized)

the median sales prices of homes within that high school boundary increases

by more than $4,000.00. The neighboring median price is also reporting an

increase of $500.00 but is not statistically significant. Given the model

results and controlling for other variables, this model concludes to a certain

degree that an increase in building ratings causes an increase in median

sales price for homes located within the boundary of that high school.


Bayesian Heteroscedastic Linear Model Gibbs Estimates (Robust) Dependent Variable = price in ths R-squared = 0.2617 Rbar-squared = 0.1169 sigma^2 = 225.7433 Nobs, Nvars = 62, 11 ndraws,nomit = 2500, 500 time in secs = 3.9000 r-value = 4 *************************************************************** Posterior Estimates Variable Coefficient t-statistic t-probability constant 7.293676 0.227799 0.820552 units sold -0.035733 -2.836091 0.006161 beds 2.599167 0.338888 0.735839 baths 10.002170 0.818715 0.416085 price in ths 0.000120 2.039667 0.045650 building ratings 4.253115 1.088219 0.280711 W*units sold -0.037027 -2.146298 0.035770 W*beds -8.466508 -0.801459 0.425927 W*baths 19.119109 1.203180 0.233481 W*days on market -0.000030 -0.415435 0.679258 W*building ratings 3.087613 0.462221 0.645540

Figure 11: 2008 - 09 Growth Regression Using “Price in Thousands” as the

Dependent Variable


Thousands” as the dependent variable. The R-squared value of the model is

very high and the building ratings are reporting a substantial influence on

median price which is statistically significant at the 95% level. For each

single increase in building rating the median sales prices of homes within

that high school boundary increases by more than $5,000.00. The

neighboring median price is not reporting as statistically significant. Given


the model results and controlling for other variables, this model concludes to

a very strong degree that an increase in building ratings causes an increase

in median sales price for homes located within the boundary of that high

school.

Bayesian Heteroscedastic Linear Model Gibbs Estimates (Robust) Dependent Variable = price in ths R-squared = 0.5731 Rbar-squared = 0.4893 sigma^2 = 153.6742 Nobs, Nvars = 62, 11 ndraws,nomit = 2500, 500 time in secs = 3.9000 r-value = 4 *************************************************************** Posterior Estimates Variable Coefficient t-statistic t-probability constant 90.508342 2.538677 0.013652 units sold -0.028514 -3.233476 0.001961 beds -10.035217 -1.981808 0.051939 baths 3.062755 0.293659 0.770000 price in ths 0.000161 3.506389 0.000851 building ratings 5.216534 2.101722 0.039648 W*units sold -0.054348 -4.395739 0.000044 W*beds -18.218041 -1.911605 0.060551 W*baths 17.389332 0.913257 0.364645 W*days on market -0.000014 -0.235198 0.814831 W*building ratings -1.407586 -0.293032 0.770476


Dependent Variable



Thousands” as the dependent variable. The R-squared value of the model is

moderate and the building ratings are reporting an influence on median price

which is statistically significant at the 60% level. For each single increase in

building rating the median sales prices of homes within that high school

boundary increases by close to $3,000.00. The neighboring median price is

not reporting as statistically significant. Given the model results and

controlling for other variables, this model concludes to a certain degree that

an increase in building ratings causes an increase in median sales price for

homes located within the boundary of that high school.


Bayesian Heteroscedastic Linear Model Gibbs Estimates (Robust) Dependent Variable = price in ths R-squared = 0.3543 Rbar-squared = 0.2277 sigma^2 = 181.1389 Nobs, Nvars = 62, 11 ndraws,nomit = 2500, 500 time in secs = 3.9000 r-value = 4 *************************************************************** Posterior Estimates Variable Coefficient t-statistic t-probability constant 29.245482 0.780814 0.437883 units sold -0.024857 -2.409284 0.018967 beds -3.794109 -0.533639 0.595499 baths 11.177924 1.059474 0.293494 price in ths 0.000020 0.479230 0.633461 building ratings 2.937780 0.850957 0.398068 W*units sold -0.037893 -2.155049 0.035049 W*beds -12.789423 -1.105703 0.273127 W*baths 33.616593 1.594523 0.115905 W*days on market -0.000013 -0.203567 0.839359 W*building ratings 1.617740 0.352568 0.725608


Dependent Variable

The first three growth regressions conclude to a certain degree that an

increase in building ratings have a positive causal relationship with median

sales price of homes located within that building boundary. Their findings are

directional and consistent although not necessarily very convincing due to

some of the low statistical values.


B. Days on the Market

The second three growth regressions presented use “Days on the

Market” as a dependent variable. These three regressions examine how

building ratings affect the amount of time it takes to sell a home after it has

been officially listed for sale.

Figure 14 shows the growth regression for 2008 - 09 using “Days on

the Market” as the dependent variable. The R-squared value of the model is

very low. The building ratings are reporting an influence on days on the

market which is statistically significant at the 70% level. What this means is

that for each single increase in building rating the median days on the

market of homes within that high school boundary increases by 4.5 days.

This could indicate that homes located within those higher performing school

boundaries are holding out for a more favorable sales price rather than

accepting the first offer presented to them. Given the model results and

controlling for other variables, this model supports that an increase in

building ratings causes an increase in days on the market for homes located

within the boundary of that high school.


Bayesian Heteroscedastic Linear Model Gibbs Estimates Dependent Variable = days on market R-squared = 0.1373 Rbar-squared = 0.0603 sigma^2 = 246.1201 Nobs, Nvars = 62, 6 ndraws,nomit = 2500, 500 time in secs = 4.2120 r-value = 4 *************************************************************** Posterior Estimates Variable Coefficient t-statistic t-probability constant 8.619534 0.650743 0.517617 units sold -0.052480 -4.415081 0.000041 beds -0.469888 -0.062087 0.950694 baths 15.210397 1.260573 0.212186 price in ths 0.000131 2.289423 0.025472 building ratings 4.508354 1.081152 0.283817

Figure 14: 2008 - 09 Growth Regression Using “Days on the Market” as the

Dependent Variable



moderate and the building ratings are reporting an influence on days on the

market which is statistically significant at the 95% level. For each single

increase in building rating the hold time of homes within that high school

boundary increases by 6.7 days. Given the model results and controlling for

other variables, this model concludes that an increase in building ratings


causes an increase in days on the market for homes located within the

boundary of that high school.

Bayesian Heteroscedastic Linear Model Gibbs Estimates Dependent Variable = days on market R-squared = 0.3222 Rbar-squared = 0.2617 sigma^2 = 241.2537 Nobs, Nvars = 62, 6 ndraws,nomit = 2500, 500 time in secs = 4.0870 r-value = 4 *************************************************************** Posterior Estimates Variable Coefficient t-statistic t-probability constant 74.875761 2.998553 0.003900 units sold -0.048830 -4.623377 0.000020 beds -10.832374 -1.837443 0.070938 baths -7.332750 -0.615435 0.540520 price in ths 0.000130 2.811422 0.006595 building ratings 6.678426 2.408640 0.018998


Dependent Variable



low but the building ratings are reporting an influence on days on the market

which is statistically significant at the 90% level. For each single increase in


building rating the median days on the market of homes within that high

school boundary increases by 6 days. Given the model results and

controlling for other variables, this model concludes that an increase in

building ratings causes an increase in days on the market for homes located

within the boundary of that high school.

Bayesian Heteroscedastic Linear Model Gibbs Estimates Dependent Variable = days on market R-squared = 0.1923 Rbar-squared = 0.1202 sigma^2 = 215.5697 Nobs, Nvars = 62, 6 ndraws,nomit = 2500, 500 time in secs = 4.0090 r-value = 4 *************************************************************** Posterior Estimates Variable Coefficient t-statistic t-probability constant 54.166922 2.528924 0.014000 units sold -0.037151 -3.468083 0.000959 beds -5.654872 -0.813520 0.419033 baths 7.577707 0.727501 0.469658 price in ths 0.000020 0.471833 0.638703 building ratings 6.019838 1.852751 0.068679


Dependent Variable

The second three growth regressions do not have the advantage of high

statistical significance; however, the evidence that they do provide


concludes that an increase in building ratings has a positive causal

relationship with median days on the market located within that building

boundary. Their findings are directional and consistent although not

necessarily very convincing due to some of the low statistical values.

C. Price per Square Foot

The final three growth regressions presented use “Price per Square

Foot” as a dependent variable. These three regressions examine how

building ratings affect the selling price per square foot of homes within their

own boundary as well as homes within the neighboring boundaries.

Figure 17 shows the growth regression for 2008 - 09 using “Price per

Square Foot” as the dependent variable. The R-squared value of the model is

moderate. The building ratings are reporting a positive influence on median

price per square foot but it is not statistically significant. The neighboring

median price per square foot is reporting a positive increase of $2.00/sf

($12.00/sf total, divided among six neighbors) and is statistically significant

at the 85% level. This means that for every single increase in building

ratings, neighboring values increase by $2.00/sf. Given the model results

and controlling for other variables, this model concludes to a certain degree

that an increase in building ratings has a relationship with median sales price

per square foot in neighboring homes.


Bayesian Heteroscedastic Linear Model Gibbs Estimates (robust SLX) Dependent Variable = psqft R-squared = 0.3316 Rbar-squared = 0.2006 sigma^2 = 437.4247 Nobs, Nvars = 62, 11 ndraws,nomit = 2500, 500 time in secs = 3.9310 r-value = 4 *************************************************************** Posterior Estimates Variable Coefficient t-statistic t-probability constant -7.259896 -0.166047 0.868661 units sold 0.032480 1.660797 0.101804 beds -4.680585 -0.486445 0.628367 baths 36.761722 2.319735 0.023663 days on market 0.180993 1.203135 0.233498 building ratings 3.615797 0.735586 0.464756 W*units sold 0.052536 2.039656 0.045652 W*beds -13.639288 -0.785643 0.435069 W*baths 4.643986 0.198525 0.843284 W*days on market 0.139686 0.634297 0.528221 W*building ratings 12.193043 1.433927 0.156617

Figure 17: 2008 - 09 Growth Regression Using “Price per Square Foot” as

the Dependent Variable


Square Foot” as the dependent variable. The R-squared value of the model

indicates that the value of prediction is strong but the building ratings are

not reporting at a statistically significant level in either own boundary or in

neighboring boundaries. Given the model results and controlling for other

variables, this model is inconclusive.


Bayesian Heteroscedastic Linear Model Gibbs Estimates (robust SLX) Dependent Variable = psqft R-squared = 0.4936 Rbar-squared = 0.3943 sigma^2 = 341.6268 Nobs, Nvars = 62, 11 ndraws,nomit = 2500, 500 time in secs = 4.0710 r-value = 4 *************************************************************** Posterior Estimates Variable Coefficient t-statistic t-probability constant 6.138559 0.125541 0.900502 units sold 0.035412 2.197447 0.031733 beds -11.677368 -1.299040 0.198741 baths 28.476023 1.795669 0.077420 days on market 0.221641 1.055440 0.295319 building ratings 1.029059 0.276474 0.783103 W*units sold 0.063967 3.582816 0.000669 W*beds -25.891583 -1.591587 0.116564 W*baths 48.272539 1.937201 0.057280 W*days on market 0.268942 1.164558 0.248659 W*building ratings -4.831280 -0.740056 0.462059




Square Foot” as the dependent variable. The R-squared value of the model

indicates that the value of prediction is very strong but the building ratings

are not reporting at a statistically significant level in either own boundary or

in neighboring boundaries. Given the model results and controlling for other

variables, this model is inconclusive.


Bayesian Heteroscedastic Linear Model Gibbs Estimates (robust SLX) Dependent Variable = psqft R-squared = 0.6091 Rbar-squared = 0.5324 sigma^2 = 341.6191 Nobs, Nvars = 62, 11 ndraws,nomit = 2500, 500 time in secs = 3.9630 r-value = 4 *************************************************************** Posterior Estimates Variable Coefficient t-statistic t-probability constant 30.299765 0.695816 0.489144 units sold 0.058590 2.929804 0.004742 beds 0.673026 0.069974 0.944439 baths 17.370940 1.145805 0.256279 days on market 0.375865 2.420423 0.018446 building ratings 1.518859 0.379158 0.705866 W*units sold 0.091619 3.394287 0.001204 W*beds -53.844350 -3.132015 0.002649 W*baths 54.246298 2.189791 0.032311 W*days on market 0.246338 1.050752 0.297451 W*building ratings -1.369239 -0.214104 0.831169



The final three growth regressions have the advantage of high R-squared

values indicating a strong level of prediction but unfortunately they were not

supported with results that were statistically significant. All three models did

indicate that an increase in test scores had positive impact on median price

per square foot; however, that indication was not supported with strong

statistical significance.


Conclusion

After controlling for the various housing factors associated with the

housing market in conjunction with building ratings, the regression model

results point toward the following housing market dynamics:

School performance has a consistent and positive association with

overall housing price, suggesting that consumers pay a premium to

purchase houses within the boundary zone of high-performing high

schools;

School performance has a consistent and implicitly negative

association with the number of days on market – that is, houses within

the boundary zones of high-performing schools tend to be on sale

longer than houses with the boundary zones of lower-performing

schools – suggesting that homeowners within the boundary zone of

high-performing high schools are aware of their favorable position and

likely reject lowball offers, holding out for an offer that is more

favorable; and

School performance has a consistent and positive association with

price per square foot, suggesting not only that consumers pay a

premium to purchase houses within the boundary zone of high-

performing schools but that consumers pay a housing premium for


school performance on houses of all sizes from small bungalows in old

neighborhoods to McMansions in new subdivisions.

In terms of overall conclusions from the statistical analysis, this initial

research indicates that school quality plays in important role in pricing the

real estate market and only a moderate role in the length of time it takes to

sell real estate. Given these conclusions, it seems clear that real estate

professionals can capture value in their housing transactions by promoting

school quality as a fundamental attribute of their housing listings and school

districts – even at the building level – can build value in the local housing

stock by maintaining and improving school quality measures.

As an initial exploratory project, this study provided some interesting

results and potentially opened the door to further research in a similar

direction. Even though the results shown did not have the benefit of

consistently high probability, they certainly were not random and did not

point in the direction of rejecting an impact. With different data and the use

of additional models future research may be able to conclude positive causal

relationships between school performance and housing market factors with a

high degree of statistical accuracy.

Moncrief yin school ratings housing factors final report

Real Estate