Factorial Design Of Analysis Research Articles

A spreadsheet heuristic to classify random survey samples for a factorial experimental design

Industrial engineers, ergonomists, production operations managers and other field researchers may be confronted with large panels of potential respondents whose characteristics will not fit into a balanced factorial analysis design. This paper describes a microcomputer spreadsheet application that allows construction of an approximately unbiased sample in a 2n factorial design, using the Plackett-Burman reduced cell model as screening device. Binary criteria for predictor variables and binary-to-decimal conversion are used as an aid to obtain the “best” respondent selection for subsequent data analysis. Two separate studies are presented to illustrate the technique.

Antigen-presenting B cells and helper T cells cooperatively mediate intravirionic antigenic competition between influenza A virus surface glycoproteins.

Parenteral vaccination of BALB/c mice primed by infection with H3N2 variants of influenza A virus results in a reduced production of N2 antibody in response to homologous (H3N2) vaccine compared with the response to an H7N2 vaccine equal in N2 immunologenicity. We now have studied the interaction in vitro of purified splenic B and T lymphocytes from variably immunized mice to ascertain the cellular basis of the hemagglutinin (HA)-influenced antibody response to neuraminidase (NA). Assay of the proliferative response of T cells in B/T-cell mixtures stimulated by H3N1 (HA-specific) and H6N2 (NA-specific) reassortant (recombinant) viruses in vitro has enabled us to differentiate cellular responses to HA and NA antigens. Using a factorial design in analysis of B/T-cell mixtures, we have shown that: (i) intravirionic HA is dominant over NA in both B- and T-cell priming; (ii) an increase in H3-specific B cells occurs in mice administered boosters of H3N2 vaccine, and an increase in N2-specific B cells occurs in those given a booster of H7N2 vaccine; and (iii) memory B cells function as antigen-presenting cells and interact with memory helper T cells in the mediation of intravirionic HA-NA antigenic competition in favor of HA. The damping of response to the NA antigen in favor of HA with reinfection prohibits balanced immunologic response to the two antigens. The present studies define further the complex immunology of influenza virus infection.

MOVEMENT IN DECLINE? OSCILLATORY MOVEMENT DISPLACEMENT THRESHOLDS INCREASE WITH AGEING*

Displacement thresholds for continuous oscillatory movement were determined for ten subjects in each of three age groups--20-23, 40-55 and 60-80 years. Oscillation frequencies of 1, 2, 3, 5, 8, 13 and 20 Hz were used, all subjects having a corrected visual acuity of 6/5. A randomized block factorial design of analysis of variance revealed significant effects of both temporal frequency of oscillation (F6,189 = 4.10, p less than 0.001) and age (F2,189 = 46.98, p much less than 0.001). The deterioration in movement sensitivity occurs equally at both high and low temporal frequencies--no significant interaction effects being revealed (F12,189 = 0.45, p = not significant). The frequency effects are ascribed to underlying mechanisms mediating movement perception. Reasons for decreased movement sensitivity with increasing age are discussed.

Cultural factors and attrition in drug abuse treatment.

The present study examined how race related to attrition in drug abuse treatment. 165 male voluntary admissions to an inpatient Substance Abuse Unit were tested at admission on functioning, motivation, attitudes, symptoms, and mood-using reliable and valid scales. One week later they completed a 12-factor scale measuring perception of the treatment milieu. Data were analyzed in a 2 x 2 factorial design of analysis of variance. There were 106 White subjects and 59 Blacks. The dropout rate for Blacks and Whites was 63%. Two factors measured at intake, motivation and social functioning, showed statistically significant interactions between race and attrition. Four ward perception factors showed a differential effect related to race and attrition. The White dropout and completer did not differ in how they perceived the ward. Blacks, however, who perceived the environment as being more insightful, spontaneous, autonomous, and practical remained. Although Blacks and Whites differed on many variables, only six showed race-related differences in attrition. Having higher motivation and poorer adjustment influenced Whites to stay. Factors in the environment, however, influenced Blacks. What might be considered a more "therapeutic" milieu was effective in helping Blacks remain in treatment. Knowledge of these cultural differences could help in designing treatment programs.