What email address or phone number would you like to use to sign in to Docs.com?
If you already have an account that you use with Office or other Microsoft services, enter it here.
Or sign in with:
Signing in allows you to download and like content, and it provides the authors analytical data about your interactions with their content.
Embed code for: Analysis 2-1
Select a size
University of Maryland School of Nursing
Part 1: The sample characteristics and baseline values, comparing the two groups’ characteristics
In part one, the data was examined and there was no missing value found. The coding was initially serial from 1 to 30 after which there were some miscoding. A descriptive analysis was conducted using excel for the baseline values of fear of falling, confidence to walk unassisted, and anxiety, for the two intervention groups; strength only and strength plus and the result obtained is displayed in table 1 below. The characteristics of the two groups described are age (years), baseline fear of falling, confidence to walk unassisted, anxiety and gender. The total number of participants were n=44 and each intervention group was assigned 22 participants (n=22). There were equal number of male and female participants assigned to each of the intervention group; male 12(54.5%) and females 10(45.5%). From table 1, the mean and standard deviation of the variables are close together, for instance, the Mean value for fear of falling is mean=38.6 (SD=4.7) for strength only group and mean=40.1 (SD=5.4) for strength plus group. Similarly, the mean and median values for age in both groups are also closely related (mean=76.5, SD=8.0, median=73.5 for strength and mean=77.3, SD=9.0, median=73.0 for strength plus group). The age range for strength and plus groups are also close together (27 and 26 respectively). From the whole table one can say that the variables are normally distributed.
Part 2: Comparing two independent group means
Null Hypothesis(Ho): There is no difference in the mean score for confidence to walk unassisted at time1 for the two intervention groups. [M1=M2]
Alternative Hypothesis(H1): There is a difference in the mean score for confidence to walk unassisted at time 1 for the two intervention groups. [M1≠ M2]
The significant level (alpha) is set at 0.05. The hypothesis is non directional because it did not state the type of difference in the mean score for confidence to walk unassisted at time 1 for the two groups. The hypothesis did not specify if it is a positive or negative difference or any other type of difference. First the data was sorted group wise (group1= strength only and group 2= strength plus. To run the t-test analysis, I first used exel to calculate the variance of confidence to walk for strength and plus group, and the variance was found to be unequal ( 15.1) and (21.6) respectively so I chose to run independent t-test assuming unequal variance. The result obtained is shown in table 2 below. The distribution of the confidence scores is approximately normal considering that the mean for the strength only group (mean=18.7, SD= 3.9) is very close to that of the strength plus group (mean=19.6, SD=4.6) (Table 2). Although table 2 did not show the values for standard deviation of confidence to work for both groups, however, the square root of variance was obtained for standard deviation for each group.
Result: An independent sample t-test was conducted to compare confidence to walk unassisted at baseline among residents (males and females) living in a retirement community. There was no significant difference in confidence to walk unassisted between the two intervention groups, strength group (M=18.7, SD=3.9) and the plus group (M=19.6, SD=4.6), t=2.02, P=0.49, two tailed (Table 2). Also, t critical (2.02) two tail is more than t statistical ( -0.70) showing no significant difference (Table 2).
Part 3: Comparing means of a single group, pretest-posttest
Null hypothesis (Ho): There is no difference between the fear of falling at baseline time 1 and the fear of falling after 90days of applying the intervention. [M1-M2=0)
Alternative hypothesis (H1): There is a difference between the fear of falling at baseline time 1 and the fear of falling after 90days. [M1-M2 ≠0]
The significance level for alpha is set at 0.05. The hypothesis is non directional because it is testing that the mean will be merely different but did not tell if the difference will be high or low, negative or positive. The level of measurement is interval/ratio. The distribution is near normal or normal distribution because the mean of fear of falling at baseline (M=38.6, SD=4.7) and that of falling after 90days (M=40.1, SD=5.4) are very closely related/ almost equal. The type of t-test to be used here is paired t-test because we are comparing variables within a single group. To get estimate of whether population variance is similar, the standard deviations (square root of variance) were compared and found to be similar. Thus the database is assumed to be of equal variance.
The scores for fear of falling at baseline (time 1) and the fear of falling after 90days were compared. The mean, standard deviation and number of observations (N) for each group was determined. Next the data was analyzed using the excel data analysis. The means of the of fear at baseline and at 90days for the subjects were compared using the t-test for paired, two samples for mean as shown in table 3.
Result: A paired sample t-test was conducted among retirement home residents to determine if there is a difference between baseline fear and fear after 90days of receiving an intervention. There was a statistically significant improvement in the level of fear in the strength plus (M=33.8, SD=6.3) from baseline strength only group (M=39.4, SD=5.1), t statistical value=8.28, P<0.0001, one tailed (Table 3) from baseline to 90days. Therefore, we reject the null hypothesis which states that there is no difference in fear in the two intervention groups and accept the alternative which states that there is a difference in fear in the two intervention groups.
Part 4: Comparing means of more than two groups.
Null Hypothesis: There is no difference in the fear of falling at 90days for the three levels of anxiety (low, medium and high).
Alternative Hypothesis: There is a difference in the fear of falling at 90days for the three levels of anxiety (low, medium, high).
The significance level alpha is set at 0.05. The standard deviation of fear scores at 90days, at the three levels of anxiety low (M=33.3, SD= 4.4), medium (M=36.4, SD=6.6), High (M=31.7, SD=7.2) are similar. The value for standard deviation was obtained by finding the square root of the variance for each group. The hypothesis is non directional because it is testing that there is a difference but did not tell us the direction of the difference whether high or low. The distribution is determined to be normal because the means of the three groups are closely related (Table 4). The type of ANOVA used is ANOVA single factor, because we are comparing fear at three different levels of anxiety.
The data for fear at 90days was sorted into three different levels of anxiety (1=low anxiety, 2=medium anxiety and 3=High anxiety). The set of observations for the variable fear2(fear at 90days) for each level of anxiety were entered into excel. The mean (M) standard deviation (SD) and number of observations (N) are as follows; low anxiety (M=33.3, SD=4.4, N=15), medium anxiety (M=36.4, SD=6.6, N=14) and high anxiety (M=31.7, SD=7.2, N=15). The data was then compared using ANOVA single factor for comparing 3 different groups as shown in Table 4.
Result: A one way between groups ANOVA was conducted to determine the impact of the mobility intervention on fear of falling among retirement home residents. There was no statistically significant difference in at least two of the group means (F=2.17
, F-critical= 3.23 P= 0.13). The mean score for the medium anxiety group was the highest (M= 36.4, SD= 6.6), followed by the mean of the low anxiety group (M=33.3, SD=4.4), while the mean of the of the high anxiety group was the lowest of the 3 groups (M=31.7, SD=7.2). Since the ANOVA test is not significant, the post hoc t-test is not needed.
Characteristics of clients at baseline (n=44)
Fear of falling
Confidence to walk unassisted
t-Test: Two-Sample Assuming Unequal Variances
Hypothesized Mean Difference
t Critical one-tail
t Critical two-tail
t-Test: Paired Two Sample for Means
Anova: Single Factor
Source of Variation
zzzzzzz ======Reference placeholder, don't delete====== zzzzzz
Running head: ANALYSIS 2 1
ANALYSIS 2 10 Therefore, we reject the null hypothesis which states that there is no difference in fear in the two intervention groups and accept the alternative which states that there is a difference in fear in the two intervention groups.