"They'll tell you that one particular lady had a wonderful experience," said Kwapnoski, "but when you've got women in the millions saying differently, it just proves that it is a widespread problem in the company. "
She would not have been able to bring the suit alone, she said.
"My voice against Walmart?" she asked. "Are you kidding me, they would sweep me under the carpet real fast. When we band together, then Walmart has to listen to us."
Lawyers for the women said that besides the anecdotal evidence, statistical analysis demonstrated that hourly and salaried retail female employees received lower pay and less advancement opportunities than their male counterparts.
"The patterns are inescapable," Sellers said. "This is not simply a case where you group together a bunch of independent decision-makers, it shows a pattern that is stunning in its consistency showing that women were uniformly disadvantaged." While there were stores across the country, Sellers alleged, Walmart managed its operations and employment practices in a centralized manner.
But lawyers for Walmart vigorously denied the allegation.
"In Walmart's retail stores, women made up two-thirds of all employees and two-thirds of all managers," Boutrous said. "The record actually shows that the number of women who were promoted at Walmart was equal to or greater than the number who applied for the jobs.
"Plaintiffs' data is deeply flawed and highly misleading," he added. "Where the data runs counter to their theory, they simply disregard it or re-label it to obscure the major gaps in their case."
Arguments at court are likely to be technical and revolve around the technical standards of class certification. Walmart argues that the lower courts were wrong to allow the women to join together in their lawsuit because they don't satisfy the so called "commonality requirement" which is necessary to prove that they can proceed as a class.
"The commonality requirement is part of the rule of civil procedure governing class actions," said Andrew Trask, co-author of "The Class Action Playbook." "It means that the women's claims have to be similar enough so that if you prove one woman's claim you have proven every woman's claim."
In court briefs, Walmart said that so many different women, working at so many different Walmarts across the country, could not bring forward any issues that are common to them all.
"Keep in mind there are 3,400 stores, 170 different job classifications, 53 different departments in the store and 41 different retail regions, each of which has its own vice president," said Trask.
Timothy Sandeful of the Pacific Legal Foundation, who submitted a brief in support of Walmart, disagreed with the lower court's reliance on statistical analysis provided by the plaintiffs to document what they said was widespread discrimination at Walmart.
"There's all sorts of reasons why people might choose to work in other jobs, or choose to not work as managers, or stay out of the workplace for a long time that are not proof of discrimination," he said.
But Sandefur's argument enrages some of the women's rights group supporting the Walmart plaintiffs.
"Statistical evidence demonstrates that even if you control for job performance and seniority, pay discrepancies existed at all the major job groups within Walmart," said Sarah Crawford of the National Partnership for Women and Families.
"These women are seeking to vindicate their civil rights, their rights to fair pay and promotions at Walmart in the face of a great deal of evidence that systemic discrimination was standing in their way," she said. "We are still trying to get our day in court."
ABC News' Susanna Kim contributed to this report.