Overview

America’s Most Trusted® surveys are conducted by Lifestory Research. Founded in 2012, the surveys stress the important influence of consumer trust in brand evaluations and purchase decisions. The core index used to evaluate and compare brands is the Net Trust Quotient Score.

America’s Most Trusted Study tracks brands that provide services or products to customers in their home. The brands included in the study are those that are among the largest in the residential new home industry. The goal of this study is not to included all brands within each category. As a result, many brands are excluded from the study.

Survey Methodology

America’s Most Trusted® surveys are conducted throughout the course of the year. The goal of administering the survey over the course of a 12 month period is to reduce any historical artifacts or bias in events occurring in a specific time period. The survey is designed to address the nuances of each brand category included in the study. Brands that serve customers across the country, as well as brands that serve customers within a specific geography, are administered to match these brand conditions.

Some categories include restrictions as to who can complete the survey. Namely, the following are restrictions used to capture the most qualified participants taking the survey: (1) Participants evaluating home builders or residential realtor organizations must indicate that they are currently shopping for a home. (2) Home building participation also requires that people have a household income of at least $40,000. (3) Home building has age restrictions based on the brand category. (4) Among the general home shopping population, someone must be over the age of 25 to participate in the study. Among the active adult resort home shopper population, people must be over the age of 50.

The America’s Most Trusted® study is designed as a multifaceted survey instrument in which all participants complete certain sections of the survey, as well as specific battery of questions that are only completed by a subsample of participants.

As an example, those people how indicate that they are actively shopping for a home are asked to complete the applicable sections of the survey that related to this sub sample characteristic. Moreover, certain battery of brand category questions are randomized to participants in order to manage the length of survey objectives in the study.

Participants in the study are shown a specific brand category list. This list typically includes the largest brands within the category being tested. (It should be noted, we make every effort to include all qualified brands, however, some brands may not be included in the study). The first question participants are asked to complete in the brand category is the identification of the brands that they have seen or heard of before (i.e., brand awareness). If a participant indicates they are aware of the brand, they are then asked to evaluate the brand along a set of attributes including trust. If a person indicates they are unaware of a brand, they do not complete any subsequent questions about that brand. In short, brand awareness is a requirement in order for people to complete questions about a brand. A brand is included in the trust rankings only if the brand awareness is significant enough to generate a reasonable sample size. Given this criterion, some brands not reported in the rankings were excluded because the brand did not garner a high enough brand awareness rating.

Sampling

The study uses a non-probability sample design in which participants are recruited via online panels. The advantages and limitations of online panels as it relates to representation of the sample to the population has received much attention from research professionals. While this debate is ongoing, in recent years the tide has begun to change and more organizations are recognizing the value and quality of online research samples. The inclusion of a margin of error is typically restricted to probability samples. However, given the preponderance of online studies noting the margin of error, here we outline the nature of margin of error as well as note the margin of error computed for this study.

To better understand the notion of sampling error, it is helpful to recall that data from a sample provides merely an estimate of the true proportion of the population that has a particular characteristic. If 100 different samples are drawn from the same sampling frame, they could potentially result in 100 different patterns of responses to the same question. These patterns, however, would converge around the true pattern in the population.

The sampling error is a number that describes the precision of an estimate from any one of those samples. It is usually expressed as a margin of error associated with a statistical level of confidence. For example, a presidential preference poll may report that the incumbent is favored by 51% of the voters, with a margin of error of plus-or-minus 3 points at a confidence level of 95%. This means that if the same survey were conducted with 100 different samples of voters, 95 out of the 100 different samples (95% confidence level) it would be expected to show the incumbent favored by between 48% and 54% of the voters (51% ± 3%).

In this study, the overall margin of error (confidence interval) at a 95% confidence level is ± 1.00%. This margin of error is applicable to each brand category. Namely, given that several thousand responses are collected within each brand category, we are able to note this margin of error for the overall study as well as for the brand specific category. This means that any of the results we find in this study are very likely to the actual responses generated if the same study were performed multiple times. With that said, it is important to understand that the margin of error changes based upon the level of analysis you are performing. In this case, when the sample size is constrained because only a certain number of people of the sample population fit into a specific criteria, the margin of error adjusts as well.

Before analysis was performed on the dataset, the sample was weighted. Sample weighting corrects for limitations in the sample that might lead to bias and other departures between the sample and the reference population. Sample weighting is used to weight the sample back to the population from which the sample was drawn. By definition, this weight is the inverse of the probability of being included in the sample due to the sampling design. Persons in under-represented get a weight larger than 1, and those in over-represented groups get a weight smaller than 1. In the computation of means, totals and percentages, not just the values of the variables are used, but the weighted values. America’s Most Trusted™ responses are weighted for age, gender and income to parameters from the U.S. Census Bureau’s 2015 American Community Survey.

Index Score Calculation

To assess the Net Trust Quotient Score, consumers are asked to evaluate how much they trust each of the brands in the study. Consumers are asked: Based on anything you have seen or heard, what is your impression of the trustworthiness of the following brands? Responses can range along an 11 point scale of Very Trustworthy to Not Trustworthy. Net Trust Quotient Scores are calculated based on how a consumer evaluated a specific brand along the 11 point scale. Scores are divided into three categories: “advocates,” customers who feel a significant trust toward the brand; “neutrals,” those who trust a specific brand, but do not see a specific brand as standing on the shoulders of other brands; and “antagonists,” who are skeptics with little, if any, trust in a specific brand.

To compute an index score, the results from the scale responses are converted into a z score. A z score is a common statistical way of standardizing data on one scale so a comparison can take place is using a z-score. Each z-score corresponds to a point in a normal distribution and as such is sometimes called a normal deviate since a z-score will describe how much a point deviates from a mean or specification point.

A z-score (aka, a standard score) indicates how many standard deviations an element is from the mean. A z-score can be calculated from the following formula: z = (X – μ) / σ, where z is the z-score, X is the value of the element, μ is the population mean, and σ is the standard deviation.

Once the z scores are identified, these scores are converted to T scores. T scores are used to inform individuals how far their score is from the mean. An advantage of using a T score over a z score is that T scores are relatively easy to understand and compare across each brand in the study.

In this study, T scores have a mean of 100 and a standard deviation of 10. All brand T scores, across all categories, are based on all brands in the study. The mean scores were standardized beginning in the 2016 study, in which the results were published in January of 2016. An index score of 100 is average for all home brands included in the study. Scores in 2016 and 2017 share the common mean score of 100. By using a category wide index allows home based brands to be compared across each category included in the consumer research study. In addition, this approach allows scores to be compared over time. Scores published in years before 2016 were not based on this criterion and as a result scores for 2013, 2014, or 2015 cannot be compared to those scores in 2016 and 2017.