Final Report Summary - SCHOOL FOOD ENVIRON. (School Food Environments)
Policies in schools can have unintended consequences. This research program evaluated how school policies related to school food environments – namely restricting student access to carbonated beverages within school or expanding dining options during lunch by allowing students off campus – generated unforeseen consequences.
For the first project of this program, carbonated beverage ban policies were documented for U.S. school districts, including the exact month these were implemented. These were matched to household-level carbonated beverage purchases for each month spanning 6 years, available from Nielsen Homescan Data. Due to highly-detailed household geographic information and demographics available for each household member in the Nielsen data, each household was also matched to their regional school district and whether they have children at the appropriate age group for the various grade-levels within the school district. This allowed me to assess whether household purchasing patterns change in response to carbonated beverage bans in their child's school.
The findings show that when a high school household member experiences a school district carbonated beverage ban, the household increases its non-diet carbonated beverage monthly consumption by roughly the equivalent of 3.4 cans. The response only among households with high school members is consistent with the notion that it is at high schools that these bans are most binding. The paper proceeds to present evidence based on statistics from the American Beverage Association that this increase is equivalent to 67-75% of the average high school student's non-diet soda consumption when carbonated beverages are readily available in school. Thus, when taking into account that the Nielsen Homescan Data does not document all potential sources of carbonated beverage purchases outside school (convenience shops, vending machines, restaurants, etc.), this is evidence that high school students are offsetting a very large part of their non-diet soda consumption in school through purchases outside of school in response to carbonated beverage bans in their schools.
The findings of this part of the project were published in the Journal of Public Economics (volume 140, pgs. 30-50, August 2016).
For the second research project of this program, open campus policies were documented for over 460 California high schools. These policies are primarily intended to provide greater dining options to students during lunch. The data collected on the policies distinguished between open campus policies that were universal for the entire student body (unconditional) and open campus policies that provided the benefit only to students reaching a minimal academic threshold. The latter policies were intended to incentivize students – particularly marginal students - to perform better in school.
The findings of this project showed that the conditional open campus policy was indeed effective in improving students’ test scores. Interestingly, the results exhibited some evidence of the unconditional open campus policy being detrimental to students’ academic performance. Thus, the unconditional open campus policy, which was merely a policy intended to affect the school food environment, ended up affecting students’ academic performance. The analysis and findings from this project were published in the Economic of Education Review (volume 54, pgs. 95-112, October 2016).
Some unintended consequences can also result from school policies concerning student health, which to some extent can also be a partial result of school food environments. This is the idea behind a work still in progress inquiring about the effect of mandatory student weight assessments – gaining greater popularity in U.S. school districts as child obesity concerns have gained momentum – intended to monitor children’s overweight and obesity and inform of their status to individual parents as well as school/school district/state administrators interested in overall student health trends. The research project will ask whether such tests increase the likelihood of student eating disorders or bullying by placing a greater emphasis on weight measures and objectives.
For this research agenda, students in California schools will be analyzed, as California schools have bi-annual universal student weight assessments within school and run an anonymous student survey concerning risky behavior, including questions related to eating disorders, self and body image, and bullying experiences. The analysis will assess whether students in different schools and school-years respond differently to the risky behavior survey questions as a function of whether their weight assessments were before or after the survey. Data on the timing of weight assessments and the timing of the risky behavior survey was already obtained for hundreds of schools over a 4-year period. However, the risky assessment surveys for the most recent year timing data was collected for has only recently become available, such that findings from this project are not yet available.
For the first project of this program, carbonated beverage ban policies were documented for U.S. school districts, including the exact month these were implemented. These were matched to household-level carbonated beverage purchases for each month spanning 6 years, available from Nielsen Homescan Data. Due to highly-detailed household geographic information and demographics available for each household member in the Nielsen data, each household was also matched to their regional school district and whether they have children at the appropriate age group for the various grade-levels within the school district. This allowed me to assess whether household purchasing patterns change in response to carbonated beverage bans in their child's school.
The findings show that when a high school household member experiences a school district carbonated beverage ban, the household increases its non-diet carbonated beverage monthly consumption by roughly the equivalent of 3.4 cans. The response only among households with high school members is consistent with the notion that it is at high schools that these bans are most binding. The paper proceeds to present evidence based on statistics from the American Beverage Association that this increase is equivalent to 67-75% of the average high school student's non-diet soda consumption when carbonated beverages are readily available in school. Thus, when taking into account that the Nielsen Homescan Data does not document all potential sources of carbonated beverage purchases outside school (convenience shops, vending machines, restaurants, etc.), this is evidence that high school students are offsetting a very large part of their non-diet soda consumption in school through purchases outside of school in response to carbonated beverage bans in their schools.
The findings of this part of the project were published in the Journal of Public Economics (volume 140, pgs. 30-50, August 2016).
For the second research project of this program, open campus policies were documented for over 460 California high schools. These policies are primarily intended to provide greater dining options to students during lunch. The data collected on the policies distinguished between open campus policies that were universal for the entire student body (unconditional) and open campus policies that provided the benefit only to students reaching a minimal academic threshold. The latter policies were intended to incentivize students – particularly marginal students - to perform better in school.
The findings of this project showed that the conditional open campus policy was indeed effective in improving students’ test scores. Interestingly, the results exhibited some evidence of the unconditional open campus policy being detrimental to students’ academic performance. Thus, the unconditional open campus policy, which was merely a policy intended to affect the school food environment, ended up affecting students’ academic performance. The analysis and findings from this project were published in the Economic of Education Review (volume 54, pgs. 95-112, October 2016).
Some unintended consequences can also result from school policies concerning student health, which to some extent can also be a partial result of school food environments. This is the idea behind a work still in progress inquiring about the effect of mandatory student weight assessments – gaining greater popularity in U.S. school districts as child obesity concerns have gained momentum – intended to monitor children’s overweight and obesity and inform of their status to individual parents as well as school/school district/state administrators interested in overall student health trends. The research project will ask whether such tests increase the likelihood of student eating disorders or bullying by placing a greater emphasis on weight measures and objectives.
For this research agenda, students in California schools will be analyzed, as California schools have bi-annual universal student weight assessments within school and run an anonymous student survey concerning risky behavior, including questions related to eating disorders, self and body image, and bullying experiences. The analysis will assess whether students in different schools and school-years respond differently to the risky behavior survey questions as a function of whether their weight assessments were before or after the survey. Data on the timing of weight assessments and the timing of the risky behavior survey was already obtained for hundreds of schools over a 4-year period. However, the risky assessment surveys for the most recent year timing data was collected for has only recently become available, such that findings from this project are not yet available.