2025 Undergraduate Basic Needs Report

by Jeremy Elvander, Miles Barry, Phuong Thy Nguyen, Erin Park
April 27, 2026

Introduction & Objectives

Basic needs analysis is a core component of university research and planning, with educational institutions regularly supporting resources and services that help students navigate college life. Institutions like UC Davis offer a wide variety of resources to its student body, with the goal of providing support and assistance to ensure a successful personal and academic experience. Basic needs vary widely, and are constantly being reconsidered and redefined. Several paradigms exist to define human basic needs. One popular model is Maslow’s Hierarchy of Needs, which describes a hierarchical needs ‘pyramid’ with physiological needs (like food, water, or shelter) at the most basic level, and abstract needs (like self-actualization) at the highest level (McLeod). Some governmental bodies, like the United Nations, define basic needs as essentials for both survival and well-being, including markers of development like health, education, employment, public transportation, and democratic participation (UN).

From a university perspective, several basic needs are of particular importance. Physiological needs, as discussed previously, are always vital, especially elements like food and housing, which are frequent dimensions where student populations face insecurity (CSAC). Other elements, like healthcare or employment, are essential for student success and physical/financial well-being. Despite the aforementioned resources that exist at institutions like UC Davis, there are still gaps in student support, especially for specific high-risk groups. These gaps, areas where students face particular insecurity, are vital to identify and understand, so appropriate actions can be taken to ensure student safety and wellbeing. Students are far more likely to be successful personally and academically if they are supported and secure physiologically, physically, mentally, and financially, making the assessment of basic needs vital for the success of academic institutions.

Project Objectives

The primary objective of this project is to assess the state of the UC Davis undergraduate student body according to several basic needs dimensions, including healthcare, housing, food, hygiene, and employment. This work aims to identify gaps in undergraduate well-being, and understand how ASUCD basic needs resources are addressing these gaps. The primary ASUCD/UC Davis resources considered were Calfresh/SNAP, the ASUCD Pantry, Aggie Compass, Aggie Aid, Aggie Reuse, ASUCD Student Health and Wellness (SHAW), and ASUCD HAUS/VESTA.

As part of this assessment, ASUCD IRL will give feedback and recommendations to the ASUCD Executive Office, Senate, and units/commissions based on survey response analysis. IRL hopes to analyze what demographic groups struggle in different basic-needs categories, and how elements of different basic-needs interact with each other.

Methodology

Data Collection

A Qualtrics survey was used to collect 351 responses between October 6th, 2025, and October 13th, 2025. Participants were recruited through two tabling sessions which took place on October 6th and October 9th. To encourage responses and thank participants for their time, each response was compensated with one cookie. 150 cookies were handed out at each tabling session. As an additional method for data collection, the ASUCD Executive Office sent an email message advertising the survey to the student body.

Data Processing

To ensure data accuracy, integrity, and quality, the distribution of response times were modeled. The resulting distribution was approximately normal, so all responses less than two standard deviations below the mean response time were excluded from analysis. The mean response time was 150.01 seconds, and the standard deviation was 63.79 seconds. Two standard deviations below the mean corresponds with approximately 2.5% of respondents, following the empirical rule. Responses that fell below the cutoff were removed to account for students selecting answers without fully reading questions, which could skew results. After processing, a total of N = 337 responses remained, reduced from the original 353. Partial responses were kept, with incomplete responses being dropped for analysis involving missing data.

Methods of Analysis

This report employs a variety of visualization and statistical techniques. Simple visualizations like bar graphs and tables were used to highlight the distributions of responses to single questions (univariate analysis), and while exploring relationships between responses to multiple questions (bivariate/multivariate analysis).

Most statistical methods used are non-parametric, meaning they carry fewer expectations about the distribution of responses to each question. These methods do not assume that the data follows a normal distribution / bell curve, so they are better for our relatively small sample size for certain groups and research questions.

These methods are also appropriate for ordinal data. Many of our survey questions prompted respondents to respond by selecting one level of a likert scale; an example of a likert scale is a five-option multiple choice question ranging from “Strongly Disagree” to “Strongly Agree”. Likert scales produce ordinal data; “Strongly Disagree” to “Strongly Agree” follows a clear order, but we cannot assume that the distance between two adjacent points is the same for every respondent. For example, one respondent’s “strongly agree” could be equivalent to another respondent’s “agree”. We have taken this into account while conducting analysis, avoiding methods that would assume equal distance between scale points.

The non-parametric methods used include:

Kendall’s Tau — used to measure the association between two variables.
The Kruskall-Wallis test — used to determine whether survey response distributions differ between three or more groups (e.g. first years vs. second years vs. third years). The Kruskall-Wallis test is used as an alternative to ANOVA when parametric assumptions are violated (as is the case with ordinal data).
Chi-Square Test — used to determine whether one categorical factor (like struggling to access mental healthcare) is related to another categorical factor (like using SHAW).
Fisher’s exact test — used to determine whether one factor (like struggling to access mental healthcare) is related to another factor (like using SHAW). This method is an alternative to the Chi-Square test when expected counts are small.
The Mann-Whitney U test — used to determine whether survey response distributions differ between two groups (e.g. first years vs. second years). The Mann-Whitney U test is used as an alternative to the two-sample t-test when parametric assumptions are violated (as is the case with ordinal, Likert-style data).

Other methods include:

Linear regression — used to model the relationship between variables and estimate how much one variable changes for each unit increase in another. This was used to calculate the average increase in SHAW awareness per increase in the start year.
Logistic regression — used to predict the likelihood of a binary outcome (like yes/no, satisfied/dissatisfied) based on one or more factors. This was used to determine which types of healthcare struggles were predictive of SHAW use, and to model the relationship between HAUS awareness and HAUS use.

All data was cleaned and analyzed using R. Packages utilized include the tidyverse library (including ggplot and dplyr), gt and gtsummary, psych, rstatix, and coin. For all hypothesis testing, a significance level of ɑ = 0.05 was used. This value corresponds to the probability of committing a Type I error, and is the cutoff used to determine the significance of a p-value.

Data Analysis

Sample Characteristics (Demographics)

All collected demographic information was optional, with respondents having the ability to skip questions should they choose to. Demographics questions were split into academic, financial, and personal identity categories.

Academic Demographics

The academic category consisted of questions on start year, college, transfer status, international status, and first generation status. Among respondents (N=319), The majority of respondents began at UC Davis in 2024 (34%). 43 students (13%) began in 2025, and 74 (23%) and 75 (24%) students began in 2022 and 2023, respectively. A total of 17 students (5.3%) started in 2021 or before.

Collected college demographics aligned well with known population demographics (https://aggiedata.ucdavis.edu/). The survey sample had 79 (25%) students from the College of Agricultural and Environmental Sciences, which is close to the true population percentage of 25%. 65 (20%) of respondents were a part of the College of Biological Sciences, and 60 (19%) were from the College of Engineering. The dominant college was the College of Letters and Sciences with 115 (36%) of responses, which is relatively close to the true population percentage of 40%.

45 (14%) of students were transfer students, and 36 (11%) were international students. Among respondents, 110 (34%) reported being first generation college students.

All academic demographics are summarized in Appendix A. Table 1.

Financial Demographics

The financial category consisted of questions on work study, financial aid, and AB540 status. 59 respondents (18%) reported participating in a work-study program, and 160 respondents (50%) reported utilizing financial aid. Respondents were additionally asked whether they utilized the California Nonresident Tuition Exemption, commonly known as AB540 exemption. This exemption allows certain students who meet residency requirements, including undocumented students, to be exempt from paying nonresident tuition at California public institutions. 18 respondents (5.6%) reported utilizing the AB540 exemption.

All financial demographic information is displayed in Appendix A. Table 2.

Personal Identity Demographics

The personal identity category consists of questions on gender, ethnicity, and race. 205 respondents (64%) reported their gender as female, with 91 (29%) reporting male. 15 students (4.7%) reported their gender identity as non-binary/other. This reflects UC Davis undergraduate population demographics closely, with 60.9% being female and 39.1% male. 68 respondents (21%) reported their ethnicity as hispanic or latino. Race demographics are similar to previous IRL survey work, with a majority of respondents being Asian (180, or 57%).

The full personal identity demographic information is displayed in Appendix A. Table 3.

Housing

Summary of Responses

Housing was considered a critical dimension of undergraduate basic needs, with four questions being designed to assess students' experiences with housing affordability and stability. Students were asked if they had experienced housing insecurity in the past year through a binary question (yes/no), with 37 respondents (11%) responding ‘Yes’ (Appendix A. Table 4.)

Bar chart: many 'Yes' (~300) vs few 'No' (~20) responses to housing insecurity — *Figure 1. Responses to ‘Have you Experienced Housing Insecurity*

The definition of housing insecurity was not explicitly stated in the survey text. For the purposes of this survey, housing insecurity is defined as the lack of safe, stable, or affordable housing, including conditions such as poor-quality housing, overcrowding, being unable to afford rent, and homelessness.

To study the financial aspect of housing insecurity, respondents were asked how often they have struggled to afford their housing, on a five-point Likert scale. The response options were ‘Never’, ‘Sometimes’, ‘About half the time’, ‘Most of the time’, and ‘Always’. This question quantifies financial housing insecurity on an ordinal scale, where responses indicating higher frequencies are associated with higher levels of housing insecurity, in an abstract sense. Among respondents (N=329), 205 (62%) said they never struggled to afford housing, with an aggregated 38% answering sometimes or more. Full responses are summarized graphically in Figure 2 and numerically in Table 1.

Bar chart: most respondents 'Never' struggled to afford housing; few reported monthly, weekly, or daily. — *Figure 2. Responses to ‘How often have you struggled to afford your housing?’*

Chart: housing affordability — Never 62%, Sometimes 27%, About half 6.1%, Most 3.0%, Always 1.8% — *Table 1. Responses to ‘How often have you struggled to afford your housing?’*

A recurring theme for comparison was how basic need dimensions interacted with each other, and how students balance limited financial resources across several basic need categories. To this end, respondents were asked if they had delayed the purchase of other basic needs (like food or clothing) to cover housing costs, in the form of a yes/no question. 96 respondents (29%) answered yes.

Graphic: Bar chart showing Yes responses much higher than No for delaying purchases. — *Figure 3. Responses to ‘Have you delayed the purchase of other basic needs to cover housing costs?’*

Table showing delayed other basic purchases to cover housing: Yes 96 (29%), No 231 (71%) — *Table 2. Responses to ‘Have you delayed the purchase of other basic needs to cover housing costs?’*

Completing the housing section of the questionnaire, respondents were asked whether they feel confident they will have stable housing for the next 6 months. This question was designed to assess what factors influence future beliefs about an individual's housing situation, and see what differences exist between current housing insecurity and future housing stability. To account for respondent uncertainty, ‘Yes’, ‘No’, and ‘Maybe’ were given as response options. Of the 327 responses collected for this question, 261 of those surveyed (80%) responded yes, 53 (16%) responded maybe, and 13 (4.0%) responded no.

Bar chart showing three response counts about delaying basic-need purchases to pay for housing — *Figure 4. Responses to ‘Have you delayed the purchase of other basic needs to cover housing costs?’*

Table: housing confidence (N=327) — Yes 261 (80%), Maybe 53 (16%), No 13 (4%) — *Table 3. Responses to ‘Do you feel confident you will have stable housing?’*

Related to housing, questions were asked about ASUCD/University resources corresponding with various basic needs categories. Regarding housing, respondents were asked about their familiarity, and use of, ASUCD HAUS/VESTA resources. Housing Advising for Undergraduate Students (HAUS) is an ASUCD unit focused on providing support for students searching for housing or experiencing housing insecurity. Vital Emergency Shelter and Transitional Assistance (VESTA), is a subunit of HAUS that supports Aggie House, a transitional housing program for undergraduate students (Basada). Familiarity was assessed on a 5 point Likert scale, with responses including ‘Not familiar at all’, ‘slightly familiar’, ‘moderately familiar’, ‘very familiar’, and ‘extremely familiar’.

Bar chart showing most respondents "Very familiar" with HAUS/VESTA; few are not familiar. — *Figure 5. Responses to ‘How familiar are you with HAUS/VESTA?’*

HAUS/VESTA familiarity chart — Not 61%, Slightly 16%, Moderate 15%, Very 3.4%, Extreme 4.6% — *Table 4. Responses to ‘How familiar are you with HAUS/VESTA?’*

A majority of students (61%) reported being not familiar at all with HAUS. In conjunction with this question, respondents were asked whether they used HAUS/VESTA resources, in the form of a yes/no question. From the collected responses, 7.7% of students (25) have used HAUS/VESTA resources.

Bar chart: large "Yes" (~325) vs small "No" (~20) responses to HAUS/VESTA resources — *Figure 6. Responses to ‘Have you used HAUS/VESTA Resources?’*

Table showing use of HAUS/VESTA resources: Yes 25 (7.7%), No 299 (92%) — *Table 5. Responses to ‘Have you used HAUS/VESTA Resources?’*

Exploratory Analysis

To prepare for targeted statistical testing, exploratory analysis was also conducted on the relationship between housing questions and key demographic variables.

Housing insecurity was relatively similar among different racial demographics, with small sample sizes for Native Hawaiian/American Indian groups contributing to higher variance in housing insecurity. For different racial groups, responses to whether one has delayed the purchase of other basic needs is statistically similar, though proportions are higher compared to housing insecurity responses across groups.

Bar chart: housing insecurity by race; light blue = No, teal = Yes; most groups report No. — *Figure 7*

Stacked bar chart showing proportions of delayed purchases for housing by race/ethnicity (Yes/No) — *Figure 8.*

Housing insecurity was considered across start years at Davis as well, with no immediately meaningful variation detected. The reported rates of housing insecurity was much higher for those starting their time at UC Davis in 2021 or before, though this demographic group represents a much smaller proportion of respondents (17, or 5.3%). The summary of housing insecurity by start year can be found in Appendix B. Figure 1.

Bivariate analysis was conducted on the relationship between different housing insecurity indicators. Relationships were considered for all housing questions, for a total of 6 relationships examined.

Kendall's tau heatmap showing correlations between housing variables, colored tiles and values — *Figure 9.*

A correlation matrix was constructed, using Kendall’s Tau as the measure of association. Kendall’s Tau is a non-parametric alternative to Pearson’s r correlation coefficient, and is appropriate for the ordinal response data. Individual statistical testing for association was conducted on specific variable pairings based on this exploratory matrix, described below. Further bivariate visualization can be found in Appendix B. Figure 2.

Research Questions and Statistical Analysis

Based on exploratory analysis, the primary research categories and questions considered were:

Studying housing insecurity based on financial demographics
1. Do groups that utilize financial resources experience more housing insecurity?
Studying experienced housing insecurity, and knowledge/use of ASUCD resources
1. Are groups that experience more housing insecurity more familiar with ASUCD housing resources?
2. Do groups that experience more housing insecurity use more ASUCD resources?
3. Do groups that are aware of ASUCD housing resources use them?

Statistical methods used for analysis include non-parametric testing for difference between groups via the Mann-Whitney U test, Chi-squared/Fisher’s Exact test for associations, Kendall’s Tau test for associations, and logistic regression for prediction of binary outcomes. To quantify statistically significant relationships, rank-biserial correlation, Chi-squared Phi coefficient, odds ratio, and Kendall’s Tau were selected for effect sizes.

Findings

Finding 1: Students that fall into three financial-need demographic categories struggle to afford housing more frequently, but are not more likely to use ASUCD housing resources.

Stacked bar chart: students on financial aid report greater housing affordability struggles.

Students that were members of financial-need subdemographics include those on financial aid, those utilizing work-study programs, and those who use the AB540 Tuition Exemption. Family-wise Mann-Whitney U tests were conducted on each group pairing (on financial aid vs not on financial aid, etc), utilizing a Holm-Bonferroni correction method to adjust for multiple comparisons. Based on the results of the tests, all financial-need groups were found to report struggling to afford housing more frequently (Appendix A. Table 5.). Of the three financial-need groups, work-study had the strongest association with more frequent housing affordability struggles, with a rank-biserial correlation of 0.38. Financial aid had a rank-biserial correlation of 0.29, and AB540 had a correlation of 0.15. All financial-need groups had statistically significant increased affordability struggles compared to respondents not in financial-need groups, with corrected p-values of 0.0003, 0.0003, and 0.0042.

After determining financial-need groups have more frequent housing affordability struggles, the relationship between financial demographic and ASUCD resource-use indicators was determined. A series of Chi-Square/Fisher’s Exact tests for association with Holm correction were conducted to determine whether there was a positive relationship between being a part of a financial-need group, and utilizing HAUS/ASUCD resources. With the previously stated significance level of 0.05, no financial demographic had a statistically significant relationship with using ASUCD resources. Financial-aid, work-study, and AB540 as indicators for using HAUS all had corrected p-values of 0.084 (Appendix A. Tables 6, 7, 8.). This finding shows at-risk groups that should be targeted for future outreach, and shows a disconnect between those who would benefit from ASUCD housing resources and those that actually use them.

Finding 2: Students that struggle to afford their housing more frequently have slightly greater familiarity with HAUS.

Stacked bar chart of HAUS/VESTA familiarity levels among housing‑insecure respondents

Determining if those that experience housing insecurity are familiar with ASUCD housing resources was seen as a vital way to determine whether vulnerable students were connecting with resources available to them. Two primary variables were compared against respondent’s ranking of HAUS familiarity: experienced housing insecurity and frequency of housing affordability struggles. Both variables present different views on housing insecurity, with frequency of housing affordability struggles being answered on a Likert-style 5 point scale, and experienced housing insecurity being a yes/no binary response. The frequency of affordability struggle question was the chosen indicator for familiarity analysis, since its ordinal scale would give more sensitivity to incremental differences compared to a binary response.

To this end, a Kendall’s Tau test rank correlation test was used to compare the statistical significance and strength of the relationship between housing insecurity and HAUS familiarity. Kendall’s Tau is a non-parametric ranked alternative to Pearson’s correlation test, and is preferred over other non-parametric association measures (like Spearman’s Rho) when there are many ties in the dataset. The calculated p-value was 0.0466, and the corresponding effect size Tau was 0.1 (Appendix A. Table 9.). This indicates a statistically significant, but weak, positive relationship. This result shows that students who experience greater housing insecurity are slightly more familiar with ASUCD housing resources. This finding represents a success in outreach to at-risk students, and shows that ASUCD should continue to market relevant resources to at-risk students.

Bubble chart: housing affordability vs resource familiarity; blue bubbles sized by frequency

Finding 3: For every one point increase in familiarity with HAUS, the odds of students using HAUS increase by 2.1 times.

Stacked bar chart: HAUS resource use increases as familiarity with HAUS rises.

To determine the usefulness of ASUCD housing resources to the student body, the relationship between familiarity with HAUS and actual use of HAUS was determined. Though it is obvious that those who use HAUS will be familiar with it, it is not immediately apparent if being familiar with HAUS is a relevant predictor of the use of it. To investigate this relationship, a logistic regression model was constructed, where the binary response variable ‘Uses HAUS’ was regressed on the ordinal predictor ‘Familiarity with HAUS’. The model constructed was statistically significant, with a calculated odds ratio of 2.09.

Table showing familiarity with HAUS: OR 2.09 (95% CI 1.56–2.85), p0.001

Based on the model output, familiarity with HAUS is a relevant predictor of use of HAUS, indicating that increased awareness can be a successful strategy to increase use of ASUCD housing resources. An odds ratio of 2.09 means that, for each one-unit increase in familiarity (on a 1-5 scale), an individual respondent's odds of using HAUS resources increases on average by 2.09 times.

Scatter plot with logistic curve showing higher probability of HAUS use as familiarity increases.