Elementary school students' anxiety of taking tests seems an essential issue for their testing performance. Either it is a positive or a negative influence for testing performance, it should become measurable before conducting a plausible research. Hence, a scale of testing anxiety for elementary school students (TAESS) ought to be developed first.

The purpose of this study was to develop a scale to measure elementary school students' anxiety of taking tests. 

Method

Participants

A convenient sample was used. Twenty four fifth graders in Taiwan were involved in this study. Among them, thirteen were male and eleven were female. They had been tutored by the same teacher for almost one semester before participating in the study.

Procedure

       Validity

A completing-sentence survey was delivered to all the participants. They were asked to complete the only one question: When I think of taking a test, I feel.... The purpose of this survey was to collect possible items for constructing TAESS, thus establishing face validity. Participants' responses were all carefully checked by the researcher. Items that were with the same meaning were seized, and one item was chosen to stand for the rest with the same meaning. All of the responses were detained. In the end, twenty four items were chosen for the initial components of TAESS.

After face validity had been established, an exploratory factor analysis technique was adopted to extract one factor in order to establish one dimentional construct validity. In the end, eight items were chosen in this second phase of validity establishment.

       Reliability

After the establishment of validity, the final edition of TAESS with eight items were used to establish reliability. Then, TAESS were delivered to all of the twenty four participants. Cronbach;s alpha indicated that internal consistency of TAESS for the present sample was high ( alpha = 0.936).

Statistical plans

The principal components analysis with varimax rotation method was employed. The rule of engenvalue greater than one was adopted for deciding which factor would be detained. Before doing analysis, the checking of sampling adequacy would be conducted. Kaise-Meyer-Olkin was suggested to be greater than 0.6. Bartlett's test of sphericity was suggested to be significant. The determinant was suggested to be greater than 0.00001 in order to avoid grave multicollinearity. For gaining as much information as possible, the pairwise method was employed to deal with missing data.

Results

Based on the above method operationalizations, the following eight items were suggested:

    1. 一想到考試就讓我感到緊張

    2. 一想到考試就讓我感到害怕

    3. 想到考試會讓我有擔心的感覺

    4. 想到考試會讓我有不敢面對的感覺

    5. 想到考試會讓我有怕考不好的感覺

    6. 想到考試就會讓我感到驚恐

    7. 想到考試就讓我感到焦慮

    8. 想到考試就讓我感到苦惱

Then, items were translated into English by simple direct translation in a source language centering approach:

    1. When thinking of taking a test, I feel nervous.

    2. When thinking of taking a test, I feel scared.

    3. Thinking of taking a test makes me feel worried.

    4. When thinking of taking a test, I dare not face it.

    5. Thinking of taking a test gives me a feeling of fear to get low scores.

    6. Thinking of taking a test makes me feel frightened.

    7. Thinking of taking a test makes me feel anxious.   

    8. Thinking of taking a test makes me feel vexed.